via Coraid EtherDrive array targets cloud storage.

Background

For many years, the total amount of storage users needed to support application and file environments just wasn’t a prime IT concern. The amount of raw storage needed to generate a given amount of usable storage was just arithmetic; utilization was known, but it didn’t really drive buying behavior. IT managers just bought more capacity when it was needed because getting the job done was all that mattered. Explosive data growth, the rapid adoption of server virtualization technology, and economic concerns have combined to change that; while getting the job done—effectiveness—is still crucial, doing it as efficiently as possible is now equally critical in most IT organizations. IT managers are looking for ways to be more productive and get more out of key applications with the same—or more likely less—budget.

ESG research indicates that a number of factors are driving IT decision makers towards more efficient storage solutions.  As shown in Figure 1, accelerating data growth, storage system costs, and increasing complexity are cited as significant challenges by IT managers.[1]

In addition to the storage challenges listed in Figure 1, ESG research indicates that reduced operational costs, business process improvement, and reductions in capital expenditures are also top priorities when making purchasing decisions.[2] As a matter of fact, compared to three years ago, 68% of respondents are significantly more aware of energy consumption and cooling requirements than in the past.  Put it all together and it’s clear that IT managers are looking for modular, cost effective storage solutions that are energy efficient and scalable.

EMC Unified Storage

EMC Unified Storage is built upon EMC’s CLARiiON and Celerra storage platforms, designed to address the most crucial cost and management challenges faced by storage and IT managers.  Both the Celerra and the CX4 offer storage efficiency technologies and comprehensive VMware integration.  All of these platforms and technologies are managed from the EMC Unisphere management interface.

EMC has recently announced a program guaranteeing that customers will require 20% less raw unified storage capacity (NAS and/or SAN) with EMC compared to other storage vendors. [3] This is based on “out of the box” best practices for configuration provided by the vendors.  This promise of savings does not take into account the compounded effect of the additional EMC storage efficiency technologies covered in this report.

EMC Unified Storage is designed to drive down consolidation costs and increase efficiency using features and attributes including:

• Fully Automated Storage Tiering (FAST), which enables sub-LUN/sub-volume automated, real-time data migration and placement on the appropriate tier with no intervention by administrators and no interruption to applications.
• FAST Cache technology, which utilizes EFD (enterprise flash drives) capacity as an extended cache pool to provide a performance boost across an entire storage system.
• Block data compression, which allows customers to compress inactive data and reclaim valuable storage capacity. Celerra has offered file level compression for some time with Celerra Data Deduplication.
• Virtual provisioning, which uses just in time capacity allocation and simplified volume management capabilities to reduce the total cost of ownership.
• High-speed, energy-efficient enterprise flash drive technology for applications with demanding performance requirements.
• Low power SATA drive technology, drive spin-down capability, and adaptive cooling to dynamically reduce energy use and improve efficiency.
• For file systems, Celerra also has the ability to tier to secondary archive devices such as Centera, Atmos and Data Domain
• Integration with VMware vCenter for high availability, data protection, and ease of use, including VM-aware storage management, vCenter plugins, Site Recovery manager (SRM) failover and failback, and VAAI (vStorage API for Array Integration). Many of these capabilities are being consolidated into a single universal vCenter plug-in known as EMC VSI (Virtual Storage Integrator).

This report documents ESG Lab hands-on testing of the EMC Unified Storage product portfolio with a focus on its ability to increase performance, availability, and investment protection as it reduces cost, complexity, and management  requirements.

ESG Lab Validation

ESG Lab performed hands-on evaluation and testing of EMC’s Unified Storage solution at multiple EMC facilities.  Testing was designed to demonstrate the simplicity and efficiency of managing and maintaining a mid-tier storage environment.

Simply Unified Storage Management

EMC Unisphere gives storage administrators the ability to manage the complete storage ecosystem with one simple web-based management tool.  Unisphere is a unified management interface for three EMC products:  CLARiiON, Celerra and RecoverPoint.  It has evolved from the CLARiiON Navisphere administrative interface and is compatible with previous releases of CLARiiON FLARE and with the current release of Celerra DART.  It also still supports existing customers’ CLI scripts.  EMC’s stated direction with Unisphere is to continue to integrate more device and information management capability into the product.  Over time, Unisphere will also provide a common communication point for broader IT infrastructure management tools like VMware vCenter.

ESG Lab Testing

ESG Lab tested Unisphere by examining the components contained in the management tool.  Unisphere can be accessed locally or through an array controller.  By setting up a domain containing Celerra and CLARiiON environments, ESG Lab was able to log in with one account and aggregate those multiple environments into a customizable dashboard.  From the dashboard view, ESG lab was able to drill down to any function using the tabs located at the top of the screen or by clicking directly on one of the links presented in the summary views.  An intuitive “Most Free Space” view is highlighted in Figure 3.

ESG Lab examined the storage pool feature that has been updated in the latest release of Unisphere.   In earlier releases, pools were used for thin provisioning of LUNs.  Release 30 introduces Thick LUNs to the pool, which can be contained in the same pool as thin LUNs.  ESG Lab observed that many of the new data services, such as FAST tiering (shown in Figure 4), are available only in pool LUNS.  ESG Lab was able to configure a pool that contained a combination of different disks including flash, Fibre Channel, and SATA, since pools contain no restrictions in terms of the types or location of drives.

Fully Automated Storage Tiering

As information continues to grow exponentially, storage systems need to be intelligent enough to automatically optimize for both performance and cost savings.   EMC FAST (Fully Automated Storage Tiering) and FAST Cache capabilities were designed with these goals in mind.   As shown in Figure 5, applications can have access to a mix of FAST and FAST Cache storage configurations in a networked environment.  FAST operates within a pool LUN, moving sub-LUN slices of data between storage drives according to activity level.  FAST Cache utilizes standard Enterprise Flash Drives (EFD) that can be dynamically added to the storage hardware, and there are no requirements for pool based LUNs.   Applications can automatically take advantage of the performance improvements as data is moved to the cache as needed.

EMC’s approach, using standard Enterprise Flash Drives (EFD), provides a flexible and scalable solution for cache.  Adding EFD’s to an array requires no scheduled outage and can scale to 2 TB mirrored R/W dynamically in a FAST Cache configuration.   Both FAST and FAST Cache utilize the same flash drive technology allowing EFD’s to be easily reconfigured for use with either solution.

Both solutions provide highly complementary performance improvements and are optimized to work together.  FAST Sub-LUN tiering optimizes capacity and lowers costs as colder data can be moved to slower and less expensive drives.  FAST Cache accelerates performance to address unexpected spikes in workloads.  FAST Cache provides greater performance returns since it examines data at a higher granularity (64k chunks) than FAST (1 GB chunks).   Since both FAST and FAST Cache utilize the same flash drive technology, EFD’s can be easily reconfigured for use as either cache, or as Tier 0 storage. EMC’s unified storage systems can support up to 64TB of FLASH drive technology.

ESG Lab Testing

ESG Lab performed several performance tests for both FAST and FAST Cache. The test bed for FAST contained a CLARiiON CX4-960 with a mix of EFD, Fibre Channel and SATA II drives. [4] Testing began on an Oracle test bed with a multi-user online transaction processing (OLTP) system that emulates a warehouse order entry application.  The capacity consumed by the Oracle database was 1.2 TB.

ESG Lab examined the configuration of FAST in a CLARiiON storage pool.  The properties of the tier were easy to configure and allowed ESG Lab to tier manually or according to a schedule.  Data was moved automatically between tiers based on activity level.  FAST moved in sub-LUN portions as a background process running within the CLARiiON.  As shown in Figure 6 , ESG Lab created a policy to move data within a 110 GB LUN between three tiers of storage (Flash, FC and SATA).  Tiering was set to manual to allow ESG Lab to run a baseline performance test first before moving the data.

ESG Lab ran a baseline performance test with only the 45 Fibre Channel disks.  The test was repeated after a FAST pool was created, adding the five flash drives and 15 SATA drives.   Performance in the form of relative transactions per minute improved significantly after data was moved to different drives in the tier, optimizing the faster flash drives for hot data and moving less frequently accessed data to the Fibre Channel and SATA drives.  Figure 7 shows transactions per minute were 108% faster after data was moved using FAST.

ESG Lab performed testing to examine the configuration and performance improvements of FAST Cache when added to Fibre Channel disk drives.   The test bed for FAST cache contained a CLARiiON CX4-960 with 45 15k RPM 600 GB Fibre Channel Drives and just eight 73 GB Flash drives.  The Fibre Channel drives were presented to a test server running Oracle on Windows 2003 SP2.  The database and server configuration were the same as the FAST test.

ESG Lab used Unisphere to configure FAST Cache, as seen in Figure 8. In this example, eight 73 GB flash drives were configured for read/write caching. Once enabled on the drives, FAST Cache operation was completely automatic.  There was no need to adjust settings or tie FAST Cache to any LUNs.

ESG Lab ran a baseline performance test with 45 Fibre Channel disks and repeated the test after enabling FAST Cache.  Performance in the form of relative transactions per minute improved significantly, 143% faster, with FAST Cache enabled, as shown in Figure 9.

Table 1 represents the drive configurations and performance results of both FAST and FAST Cache testing performed by ESG Lab. 

What the Numbers Mean

• The capacity of an all-Fibre-Channel database environment was doubled using five 200 GB EFD drives and 15 2 TB SATA drives as OLTP performance was increased by 108%.
• An Oracle OLTP environment saw performance increase more than 143% using FAST Cache without adding any additional drives.
• Latency, which translates directly into delays experienced by users, reduced by 65% with flash drives using both FAST and FAST Cache.

In addition, ESG Lab audited results from a test performed with SATA drives used in place of Fibre Channel.  The results showed the same performance gains as the Fibre Channel tests.  Since FAST Cache offloads most of the IO to flash drives, it minimizes the reads/writes performed against the physical disks behind cache.  This enables SATA drives to deliver performance numbers comparable to Fibre Channel configurations.

Overall, ESG Lab recorded improvements in both database transactions and response time with FAST and FAST Cache.  Specifically, database transactions increased 108% with FAST and 143% with FAST Cache.  Response times for both FAST and FAST Cache improved by 65%.

As seen in Table 2, an ESG Lab audit of cost of CX4 infrastructure indicates that a 54 TB configuration using EFD, FC, and SATA drives with FAST costs less to acquire than a 54 TB FC only configuration, provides better performance and consumes less power.

Optimizing Virtual Server Environments

CLARiiON and Celerra storage systems support VMware’s vStorage API for Array Integration (VAAI) for vCenter integration. This enables VMware administrators to easily provision and manage virtual machine storage. Improved integration hides the complexity of the underlying storage systems, but gives VMware administrators the ability to perform advanced administrative tasks including cloning volumes, creating pools and compression.

ESG Lab Testing

ESG Lab tested the integration of CLARiiON and Celerra in a vCenter environment by examining the functionality of the plugins for CLARiiON and Celerra in vSphere.  Using the navigation pane in vSphere, ESG Lab was able to right click on a VMware cluster and select EMC Celerra as one of the configuration options. From that selection, ESG Lab could quickly and easily provision storage from an available storage pool for a VMware cluster node as shown in Figure 10.

ESG Lab was also able to navigate to a VMware image and select an option under the EMC Celerra plugin to create a full or fast clone of the image.  During the cloning process, ESG Lab was able to create a storage pool to use for the new image.

ESG Lab also examined the plugin for CLARiiON with vSphere.  The CLARiiON plugin presents a tab available directly from the vSphere home page.  ESG Lab was able to log in to a vSphere environment and, from the CLARiiON page, provision new datastores and delete existing datastores.

In addition, ESG Lab looked at EMC’s new unified vCenter plugin which is planned for general availability when the next version of VMware vSphere is released.   Managing a storage environment was more intuitive as all the tasks are included as a tabbed page, much like the Unisphere interface.  ESG Lab was able navigate to a virtual machine and see the storage configuration for that image.  Figure 11 shows the EMC Storage tab available from the VMware image configuration page.

ESG Lab also examined VMware-aware Unisphere, which allows a storage administrator to look into a virtual infrastructure.  Unisphere talks directly to ESX APIs and provides end to end mapping from ESX, to LUN, to virtual machine mappings.  ESG Lab was able to monitor the state of storage requirements, including alert messages from ESX when storage is full.

Storage Efficient Compression

EMC’s block data compression feature allows administrators to compress inactive data in live volumes and reclaim storage capacity. This builds on the capacity reduction provided by Virtual Provisioning (a.k.a., thin provisioning).  Virtual Provisioning frees up allocated but unused LUN capacity that can be returned to the common storage pool. Compression provides additional benefits as it recognizes repeating patterns in allocated and used application data.  Compression is supported on both full and thin provisioned LUNs.  If the LUN is fully provisioned and part of a RAID group, the LUN is migrated to a storage pool at the same time it is compressed and the free space is returned to the RAID group.  Fully provisioned LUNs that are compressed are automatically converted to thin provisioned LUNs.

ESG Lab Testing

ESG Lab tested the results of compression on a CLARiiON by selecting a 100 GB LUN with 75 GB of mixed data types in an NTFS file system.   The data set was derived from production EMC IT department’s data and included a mix of media, Office, binary, and text files.  ESG Lab configured compression by navigating to the compression tab under the LUN properties page.  The compression properties were simple to configure and included the option to specify the rate of compression depending on the requirements for availability during the compression process.  ESG Lab chose to use high compression for this test.

Compression of a 100 GB LUN completed in 1 hour and 40 minutes.  Examining the results of the compression, as shown in Figure 14, ESG Lab observed that the LUN capacity was reduced from 100 GB to 54 GB, a savings in space of almost 50%. As expected, after compression, the LUN had been converted from a thick to a thin LUN.

ESG Lab Validation Highlights

• ESG Lab confirmed that Unisphere provides a simple and intuitive management interface for multiple EMC storage systems.
• vCenter plugins for CLARiiON, Celerra and RecoverPoint were used to perform routine storage management tasks from a VMware management console.  A unified plug-in, with planned availability in the next version of VMware vSphere, provided an intuitive interface for centralized storage management.
• With simple “set it and forget it” compression capability, ESG Lab observed a 46% reduction in CLARiiON disk capacity for common office files collected from EMC’s production IT environment.
• A 108% performance increase was recorded using FAST Sub-LUN tiering.  These results were consistent with both Fibre Channel and SATA drives.
• A 143% performance improvement was recorded with FAST Cache as compared to traditional Fibre Channel storage.

Issues to Consider

• The current VMware vCenter plugins for Celerra, CLARiiON and RecoverPoint have difference in capabilities and a different look and feel. The unified plugin that ESG Lab tested (planned for generally availability when the next version of VMware VSphere is released) has an intuitive and consistent look and feel.
• While the energy efficiency of the CLARiiON product line has been dramatically improved, full policy-based drive spin-down could be used to magnify the savings (currently, drives spin down after 30  minutes of disk inactivity)—particularly for large near-line archives that are infrequently accessed.  While this technology is currently available in the CLARiiON CX4 and EMC Disk Library product, it is not yet supported within the Celerra line.  EMC has advised ESG Lab that policy-based drive spin-down is planned for a future code release.
• While the management of EMC’s block and file storage tested by ESG Lab for this report is unified, the underlying hardware and software architectures are still separate and distinct. To be fair, EMC has made tremendous strides toward truly unified storage architecture, based on discussion with EMC, ESG Labs believes unification of midrange storage is on the roadmap.

The Bigger Truth

IT professionals are being tasked with justifying storage strategy even as senior managers struggle with budgets hamstrung by financial crisis.  How can organizations keep pace with capacity growth of 50% or more annually while staying within budget?   How will storage investments be protected—now and in the future?   How can more capacity be managed with better performance and service levels with existing staff?  Will storage investments complement—or complicate—virtual server consolidation initiatives? How will IT create a winning strategy that works for both the team and the organization?

ESG is not only impressed with EMC’s ability to continually increase the capacity, performance, and capabilities of its full product line, but also by its focus on continuously improving manageability.  In 2008, ESG Lab examined the fourth generation CLARiiON CX4 and confirmed that EMC stuck with its game plan of continuously simplifying routine management tasks, even as a number of innovative technologies were introduced into the CX4 product line which lowered the cost of ownership for CLARiiON customers.

EMC Unified Storage builds on EMC’s heritage of providing a rich set of comprehensive availability and protection options, adding advanced storage efficiency and integrated, consistent management for both SAN and NAS storage.

ESG Lab was able to manage storage area network (SAN), network-attached storage (NAS), and storage efficiency technologies simply and intuitively using the EMC Unisphere storage management interface. VM-aware storage management, vCenter plugins, and VAAI provided comprehensive storage management from within the vSphere management environment.

While ESG lab confirmed that EMC has made significant advances with the unified management of CLARiiON, Celera and RecoverPoint EMC still offers multiple separate architectures for block, NAS, and object storage. EMC has been listening to their customers and has been making significant progress toward a comprehensive, unified architecture starting at the component level, using the same disks, shelves, and controllers in the CLARiiON, Celerra, VMAX and VPLEX product lines. While there is still work to be done to get the rest of the way there, EMC has taken great steps in the right direction.

EMC Unified Storage includes a number of innovative technologies which lower the cost of ownership for EMC customers. ESG Lab used block data compression to compress inactive data and reclaim valuable storage capacity.  FAST Cache leverages enterprise flash drives to avoid the wasted cost of over-provisioned disk capacity for applications with extreme performance requirements. ESG Lab hands-on testing has confirmed that FAST sub-LUN automated tiering can use a combination of EFD, FC, and SATA drives to provide exactly the same capacity as a large pool of Fibre Channel drives at a lower cost of acquisition and with better performance.

With EMC Unified Storage, EMC has dramatically enhanced simplicity, performance, and storage efficiency for consolidated, virtualized environments. While the speeds and feeds are impressive, ESG Lab is most impressed by the continuous improvements in manageability and the long list of valuable new capabilities that have been built into the offering.  From the speed and automated operation of FAST to the dramatic manageability improvements of Unisphere and VM-aware technology, the breadth and depth of the features built into EMC Unified Storage can be used to meet the precise needs of any organization. If your organization is struggling to keep up with data growth while providing ever higher levels of performance and availability with stagnant or shrinking budgets, ESG Lab recommends that you consider EMC Unified Storage as the foundation for your own winning storage strategy.

Appendix

[2] Source: ESG Research Report, 2010 IT Spending Intentions Survey, January 2010.

[3] See EMC 20% guarantee and efficiency calculator, http://www.emc.com/products/unified-storage-guarantee/index.htm?Pid=prod_tech_unified-unifiedguarantee-070610.

[4] Configuration details can be found in the Appendix.

[5] Published list pricing data used to calculate savings can be found in the Appendix

[6] Source: ESG Research report, The Impact of Server Virtualization on Storage, December 2007.

Background

The management of file-based, or “unstructured,” content (i.e., multimedia files, Web pages, office productivity documents, etc.) has become one of the most pressing and persistent challenges facing today’s IT organizations; IT managers must store, deliver, and manage large volumes of unstructured data while meeting increasingly demanding service levels. ESG research indicates that the majority of end-users currently using or considering scale-out NAS solutions are most concerned with improving management efficiency, scalability, performance, and availability.[1]

Fujitsu Scalable File Server Solution or UDS File Services

Fujitsu Scalable File Server Solution (UDS in North America) File Services offerings, shown in Figure 2, offer a broadly scalable NAS solution that can address the unique challenges of scaling out NAS services for businesses of all sizes.

Fujitsu brings a number of economic benefits associated with scale-out NAS platforms to the table with its Symantec FileStore based solution. Scale-out NAS architectures have a number of cost advantages over scale-up solutions, ranging from start up costs to managing technology refreshes—and many steps in between. Scale-out NAS carries a lower overall cost compared to scale-up systems for a number of reasons:

• Ability to scale capacity without scaling headcount. With Fujitsu Scalable File Server Solution, it is just as easy to manage a clustered storage system with sixteen nodes as it is to manage one with two nodes. Scale-out file storage systems enable this through clustering and a global namespace, which provides a single point of management for massive amounts of file data.
• Low entry cost. The entry cost for scale-out systems varies depending on the minimum configurations supported. Fujitsu Scalable File Server Solution starts as small as two nodes and scales up from there. With clustered scale-out systems, you can add resources and scale as needed, online.
• Just-in-time scalability. Because of the modular nature of scale-out systems, there is no need to buy (and power or cool) frames, power supplies, and mostly empty cabinets in advance of storage capacity needs.
• Higher utilization rates. Because all of the NAS heads in FileStore clusters can address the entire pool of usable capacity, no capacity is locked away behind underutilized NAS heads—a common problem in scale-up systems. It is not unusual to see utilization rates of 30% or less in scale-up systems and 60% or more in scale-out systems.
• Reduced change management planning cycles. The modularity and scalability of scale-out NAS allow for extremely fast provisioning. Fujitsu Scalable File Server Solution is plug-and-play; add a storage or processor node and the system self-discovers and expands the file system or incorporates it into load balancing algorithms on the fly. There is typically no disruption of service, nor is there a requirement to plan data layouts, create LUNs, or perform data migration.
• Non-disruptive technology refresh. The process of managing technology refreshes with Fujitsu Scalable File Server Solution is faster and easier than with monolithic NAS because the cluster maps logical mount points to physical mount points in a virtualized manner, allowing back-end technology changes to be made with little or no disruption to client access.

Fujitsu Scalable File Server Solution offers advanced features still fairly new to scale-out NAS, with built-in support for storage tiering and solid state disk. Dynamic Storage Tiering allows for policy-based movement of files across storage tiers. Based on activity or desired performance levels, data can automatically be demoted to lower performing, bulk-storage tiers, including MAID, or promoted to primary tiers if it suddenly becomes very active. Consider the benefits this capability brings to environments with tiers that include solid state disk (SSD). For example, a Fujitsu Scalable File Server Solution customer can initially choose to store data in a secondary tier (SATA or Nearline SAS) and then promote it to a primary tier based on IO access. Dynamic Storage Tiering can then push the data back to the secondary tier based on IO inactivity, eliminating the need to peg data in valuable SSD real estate and then manually monitor and move it. All of this storage tiering and data movement occurs completely in the background, transparently to the end-user. The locations of the files, and even their node numbers, do not change.

This report examines Fujitsu’s Scalable File Server Solution with a goal of evaluating ease of administration, performance, and scalability as well as advanced enterprise-class features such as file system mirroring, replication, and Dynamic Storage Tiering.

ESG Lab Validation

ESG Lab performed hands-on evaluation and testing of the Fujitsu Scalable File Server Solution at Fujitsu’s Sunnyvale Campus. Testing was designed to evaluate manageability, performance, scalability, and availability. The test bed, shown in Figure 3, was comprised of the Scalable File Server Solution’s entry-level NAS storage system with two cluster nodes. Windows and Linux workstations were utilized as CIFS and NFS clients.

Ease of Deployment and Flexible Management

ESG Lab first examined the management interface of Fujitsu’s Scalable File Server solution, evaluating the ease of use and functionality of creating storage pools, file systems, and shares for CIFS, NFS, and FTP clients. Fujitsu utilizes the browser-based Symantec FileStore management tool to provide a full featured interface for all file service and storage functions. The Fujitsu Scalable File Server Solution ships with a storage array and server cluster preconfigured, so administrators can create shares and start providing file services immediately.

ESG Lab Testing

The first step in the setup of the Fujitsu Scalable File Server Solution platform was to configure TCP/IP networking using a CLI connection directly attached to the system console. ESG Lab configured the network IP address for each physical interface along with the virtual IP addresses associated with the physical IP. A console IP address was also assigned at this point. Figure 4 shows the TCP/IP configuration wizard in the CLI.

Once the IP addresses were assigned, all management was performed using the browser-based management tool. A Web browser was pointed to the same IP address that was used for the console CLI. A local administrator username and password were entered. The administrator can be a local user account or one contained in an Active Directory domain or LDAP source.

The home page of the management interface, seen in Figure 5, shows the status of the Fujitsu Scalable File Server Solution. All NAS and file system functions are managed from this page.

From the File System screen, ESG Lab selected the option to create a new file system. The selection started a two-step wizard, shown in Figure 6 that allowed ESG Lab to configure the size, layout, and storage pool location for the file system. ESG Lab created a 15 GB file system named “ESG1” with a striped layout in pool DG0001.

Next, ESG Lab clicked the shares tab in the System Status window and selected the option to add a CIFS share. As seen in Figure 7, one dialog box was displayed where the file system to share was selected, and the name and properties of the share were set.

ESG Lab tested access by mapping the network share to a network drive from a Windows client; ESG1 was mapped to the Y: drive. Figure 8 shows files and folders successfully copied to the share.

This test was repeated using the same file system, ESG1, to create an NFS share. As before, the shares tab in the system status screen was used to access the add share wizard.  Once the NFS share was exported, the share was mounted by a Linux client and files were copied into the NFS share successfully. The entire process, creating a file system and sharing it out to both NFS and Windows clients, took less than five minutes.

Next, ESG Lab used a snapshot to simulate the recovery of an accidentally deleted file. Single, manual snapshots can be created or a schedule can be set to take recurring snapshots of a file system on a periodic basis. ESG Lab created a snapshot schedule as seen in Figure 9.

Once created, the snapshot was made visible to Windows clients using the same add share wizard that was used to create CIFS shares, seen earlier in Figure 7.

Connectivity to the snapshot was tested using a Windows client. First, a file was deleted from the ESG1 share and then ESG Lab copied the file from the snapshot and back to the ESG1 share.

Performance and Scalability

One of the key benefits of Fujitsu’s Scalable File Server Solution is its ability to provide a high-level of aggregate capacity and performance that scales in near-linear fashion as nodes and storage are added to the cluster. With a maximum file system size of 256 TB and total addressable storage space of 2 PB, Fujitsu’s Scalable File Server Solution can scale to meet the needs of applications that require high performance, high capacity, or a combination of both. As additional cluster nodes are added, performance increases with the additional processing power, memory, and bandwidth of each node. As additional disks and disk arrays are added, the capacity—and performance—of the solution increases as well.

Fujitsu Scalable File Server Solution also provides advanced features that enhance the scalability of the system. Not only can file systems be expanded non-disruptively, they can also be shrunk to reclaim space for other purposes.

ESG Lab Testing

ESG Lab tested both expanding and shrinking a live file system mounted by an active client. These functions were accessed from the File Systems tab in the System Status window. ESG Lab first expanded the 15 GB ESG1 file system by 10 GB, to 25 GB total, as shown in Figure 11. After expanding the file system, files were copied into that file system to validate connectivity. The client was still connected to the share and was able to copy files with no issues.

ESG Lab then used the same wizard to shrink the file system down to 20 GB. Connectivity was tested by again copying files into the now smaller file system. The size was verified by looking at the properties of the associated share from the Windows client and confirmed: the new file system was 20 GB, adjusted with no disruption to connectivity or remounting of the share.

ESG Lab next tested the ability to grow the cluster non-disruptively.  Starting with a three node cluster, a scripted copy operation was started from a client to generate a continuous load as a fourth node was added. First, the physical IP address for a pre-wired and staged new node was added to the cluster configuration and a group of virtual IP addresses were assigned to the new node as shown previously in Figure 4.  Once the IP addresses were assigned, the node was rebooted and F12 was pressed to initiate a network boot. FileStore software was then automatically installed on the new node. Once the installation was complete, the new node appeared on the Cluster page of the FileStore management GUI as an available node, as shown in Figure 13.

At this point, adding the node to the cluster was accomplished by selecting the IP address and clicking “Add Node.”  Within a few seconds, the node was added to the cluster. While IO paused momentarily, as seen in Figure 14, it resumed with no loss of connectivity and no remount required.

SPECsfs2008 Results

ESG Lab audited published results of the SPECsfs2008 industry standard benchmark suite maintained by the Standard Performance Evaluation Corporation (SPEC). SPECsfs2008 testing measures file server throughput and response time, providing a standardized method for comparing file server performance across disparate vendor platforms. SPECsfs2008 results summarize the server’s capabilities in the context of both the number of file operations that can be handled per second, in addition to the overall latency of the file operations. While Fujitsu has not published results for a Fujitsu Scalable File Server Solution configuration as of this writing, there is a result for the 12-node Symantec FileStore Cluster ESG evaluated for this report.

SPECsfs2008 results are audited by the Standard Performance Evaluation Corporation and peer reviewed to ensure consistency. Full configuration data for each SPEC benchmark result are publicly available for download and review.[2] While this can be useful for comparison between vendors, it is important to note that not all vendors participate and publish results.

As seen in Figure 15, the 12-node cluster provided an excellent result of 176,728 SPECsfs2008 requests per second with an average response time of 4.1 milliseconds. Response time is an extremely important component of SPEC results as this is the delay that an application will experience (and pass on to users) when a storage system is stressed to its limits. A system servicing 176,728 SPECsfs2008 IO requests per second with a response time of 4.1 ms is exceptional.  In fact, FileStore has posted one of the highest SPECsfs2008 NFS result as of this writing.

Availability and System Protection

Central to the functionality of Fujitsu’s Scalable File Server Solution is high availability. The system is designed to ensure that the file system remains functional at all times—even in the unlikely event of a software or hardware failure. Clustering technologies are used to eliminate unscheduled downtime and protect data from:

• Server failures (e.g., failed processor or power supply).
• Storage failures (e.g., failed drive or controller).
• Network failures (e.g., failed network cable or NIC).

In addition, the ability to mirror the file system provides additional redundancy to reduce the need to recover the file system if a storage array is taken offline. Similarly, file system replication provides the capacity to copy entire file systems to an off-site location over a wide area network to protect against site outages. Figure 16 shows an overview of data availability options available with Fujitsu Scalable File Server Solution. In addition, anti-virus software is built into the cluster and can be activated with a license key with no installation required. Policies for anti-virus scanning can be set up individually per share. These policies can be set to scan shares on a schedule or set to auto protect mode which scans on file reads.

ESG Lab Testing

ESG Lab validated the failover capabilities of a Fujitsu Scalable File Server Solution by injecting faults into a two-node cluster. The first scenario ESG Lab tested was a network failure. ESG Lab used a client connected to the IP address on port eth1 on Node2 to continuously copy files from one file system to another. A network failure was simulated by removing the Ethernet cable from port eth1 on Node2 while the file copy was in progress. The virtual IP addresses assigned to port eth1 were moved to other available physical addresses in the cluster and the copy traffic moved to port eth2 transparently, with no administrator intervention, after only a brief pause. Figure 17 shows IO activity using Windows Task Manager on the client.

The second test simulated a storage failure by running a firmware upgrade on one controller in the ETERNUS DX80 array. ESG Lab used the same client running the same script to continuously copy files while the controller firmware was upgraded. As before, a brief pause in IO was observed while the ETERNUS DX80 failed all disk volumes over in the array. No loss of connectivity to the file system was experienced by the client during failover of the controller.

Finally, ESG Lab simulated a cluster node failure. The test was executed on a four-node cluster configuration. A client accessing two shares on Node4 was used to continuously copy files from one file system to another. ESG Lab then deleted Node4 from the cluster, at which time the virtual IP addresses originally assigned to Node4 were moved to physical IP addresses on the remaining three nodes. While the client experienced another brief pause in IO, the copy operation continued with no action required by the client or the administrator.

Fujitsu Scalable File Server Solution also allows administrators to add mirrors of a file system; this is in addition to the disk level protection in the ETERNUS array. File System mirrors can be used to enhance availability or facilitate migration of a file system to new back end storage non-disruptively. ESG Lab added a mirror of the ESG1 file system using a one step wizard. The wizard allowed the selection of disks for the mirror from all available storage. After clicking “OK,” the cluster began mirroring the file system to the new disks. Figure 19 shows the File System Details page after mirroring completed, confirming that there were now two mirrored copies of the file system.

Finally, ESG Lab tested the functionality of replication, including the ability to copy only changed blocks rather than whole files. A replication job was set up to replicate the file system cifs0, owned by a two-node cluster, to a file system on a four-node cluster. A total of 25 200 MB files were copied to cifs0 and a replication event was triggered. When the replication event was complete, the target file system, cif0t, was examined and the files were all transferred successfully, for a total of 5 GB of data moved. Next, ESG Lab renamed all 25 files in cifs0 and triggered a second replication event.

System logs showed minimal replication traffic to cif0t and the files at the target site, seen in Figure 20, reflected the new names less than a minute after triggering the replication, confirming that replication only transferred the changed metadata.

ESG Lab also tested the anti-virus feature built into Fujitsu File Services using two files with well known virus signatures. The files were copied into a share called virustest. ESG Lab then started a scan of the virustest share and the two infected files were discovered and moved to a quarantine area, shown in Figure 21.

The second anti-virus test ESG Lab performed reviewed autoprotect mode, where files are scanned on demand when read by clients. ESG Lab copied the same two infected files into the virustest share. When an attempt was made to read the files with a client, the files were automatically scanned and quarantined and access to the files was denied.

Dynamic Storage Tiering

Fujitsu Scalable File Server Solution can use multiple tiers of storage to meet a variety of application performance needs. SSD, Fibre Channel, or SAS drives are supported for performance-sensitive applications and workflows including high performance computing and rich media editing and delivery. Affordably dense Nearline SAS or SATA drives are supported for capacity-intensive applications including deep archives, backup to disk, and large scale consolidation of general purpose legacy file systems.

Fujitsu Scalable File Server Solution provides Dynamic Storage Tiering to provide automatic, hands-off movement of files between tiers based on access patterns. Policies can be set to determine how frequently the system checks for file access and what parameters are used to trigger movement between tiers.

ESG Lab Testing

ESG Lab tested the functionality of Dynamic Storage Tiering on a Fujitsu Scalable File Server Solution cluster that contained a primary tier of high speed SAS disks and a secondary tier of low speed Nearline SAS  drives.  A 100 GB file system, hscifs0, was set up with both tiers and access policies were set. The primary tier was seeded with files totaling 11 GB and the secondary tier was seeded with files totaling 54 GB. Figure 23 shows the utilization of each tier in the File System Details page of the management GUI.

ESG lab then modified 4 GB of data files residing in the secondary tier. After the scheduled scan, the changed data was moved from the secondary tier to the primary tier according to the configured policy. The two tiers were again examined in the File System Details page and primary utilization had increased to 15 GB while secondary utilization decreased to 50 GB, confirming that the 4 GB of data files that were modified had been automatically moved to the higher tier.

ESG Lab Validation Highlights

• Fujitsu Scalable File Server Solution was configured and serving files to Windows and Linux clients less than five minutes after sitting down at the console.
• The Web-based administration console can be accessed from any browser. No separate installation of an administration program is required.
• Snapshots were easy to configure and use to recover files.
• Growing the cluster by adding a new node was a straightforward, non-disruptive process as was growing and shrinking file systems.
• Audited FileStore performance was impressive, posting impressively high throughput with a very low overall response time.
• Advanced data protection and recovery capabilities typically associated with enterprise-class NAS appliances were tested. Local snapshots were used to recover files after common errors and the system stayed online through multiple simulated hardware errors, such as network outages and storage controller and power failures.
• Dynamic Storage Tiering provided automatic, hands-off movement of files between tiers based on access patterns.

Issues to Consider

• While Snapshots were easy to manage and use, administrators must manually share snapshots to allow users to recover files.  A feature to automatically share snapshots with pre-defined naming conventions and integrate with Microsoft’s Volume Shadow Copy Service (VSS) would be a useful enhancement.
• When prompting the administrator to select disks for mirroring a file system, the FileStore GUI did not exclude disks already in use by the source file system.  While the system would not create the mirror on disks already in use by the source, their appearance in the selection screen could be confusing to administrators.
• The version of SAMBA currently in FileStore can only share out a file system from one IP address, which can limit the total performance available to any single SMB share. This issue is corrected in the latest version of SAMBA, which is being integrated into FileStore as of this writing.

The Bigger Truth

The massive growth of file data flooding data centers today—and the wave that will be generated as more and more cloud storage and rich media applications come online—can easily overwhelm traditional scale-up NAS solutions. A scale-out NAS solution from Fujitsu offers scale beyond that which can be attained with traditional NAS solutions: users can start small with a two node system and affordable back-end storage and grow to a massively parallel system with Enterprise class ETERNUS arrays. The performance ceiling is raised by adding more processors and capacity is increased by adding more storage, enabling “just-in-time” scalability. And management is simple because the Fujitsu Scalable File Server Solution scale-out NAS solution is managed as a single entity—no matter how large it gets.

Legacy scale-up NAS solutions face a number of challenges. First, scale-up systems typically have capacity limits in the range of tens of terabytes, with individual file system limits between 2 and 16 TB. As capacity is scaled and limits are hit, more discrete systems are needed—and those systems need to be managed. Second, scale-up systems have fixed performance ratios; there is a fixed number of NAS heads that can be included in a single file system, typically one or two. Third, scale-up NAS has a relatively expensive price/performance ratio compared to scale-out.

A scale-out NAS solution from Fujitsu combines the field-proven performance and scalability of Symantec FileStore software, Fujitsu PRIMERGY industry standard servers, and Fujitsu ETERNUS DX disk storage system to cost-effectively address both scale-up and scale-out NAS challenges. What’s more, it’s surprisingly simple to deploy. ESG Lab was accessing files less than five minutes after getting started. It was also easy to manage via an intuitive graphical user interface.

Additionally, it supports enterprise-class NAS features that are often missing in scale-out NAS solutions; ESG Lab tested snapshots, file system mirroring, remote replication, and automated online migration between different tiers of storage. Last, but not least, it is fault tolerant and fast.

Three things in life are guaranteed: death, taxes, and information growth. Information growth will continue even in a down economy. This deluge of file-based information must be dealt with. Scale-out storage is the wave of the future—it is a path for IT managers to meet their number one storage challenge: keeping pace with overall data growth.

Because of these considerations, more and more enterprises are taking a serious look at scale-out NAS solutions like Fujitsu’s Scalable File Server Solution—clustered scale-out solutions are going mainstream. But commercial enterprises are not just interested in the increased bandwidth scale-out solutions bring to the table: users are expanding use cases for scale-out NAS thanks to the higher scalability and manageability of these systems. In short, scale-out makes economic sense.

Using an appliance-based approach that is fault tolerant and centrally managed, Fujitsu has harnessed the field proven power of ETERNUS and PRIMERGY hardware to Symantec’s FileStore software to create an enterprise-class scale-out NAS solution that is extremely scalable, extremely fast for a wide variety of applications, and extremely easy to deploy and manage.

Appendix

[2] http://www.spec.org/sfs2008/results/

Introduction

Organizations of all sizes are struggling to meet the conflicting challenges associated with information storage growth and complexity juxtaposed with global financial uncertainty. A growing number of IT managers are turning to virtualization and consolidation technologies to meet these challenges.

Background

ESG research indicates that a number of factors are driving IT decision makers toward more cost efficient storage solutions.  As shown in Figure 1, accelerating data growth, storage system costs, and increasing complexity are cited as significant challenges by IT managers.[1]

In addition to the storage challenges listed in Figure 1, ESG research indicates that reduced operational costs and reductions in capital expenditures are also top priorities when making purchasing decisions.[2] Put it all together and it’s clear that IT managers are looking for modular, cost effective storage solutions that are both efficient and scalable.

Coraid EtherDrive SAN

Coraid EtherDrive products combine commodity hardware, lightweight Ethernet networking, and a scale-out virtual storage architecture that can grow from a single appliance to multi-petabyte installations. As seen in Figure 2, Coraid provides both cost/capacity optimized and performance optimized storage appliances supporting SATA, SAS, and SSD drives. Coraid systems support all standard RAID types including RAID 0, 1, 5, 6, and 10.

To make the offering as turnkey and simple to deploy as possible, Coraid also offers HBAs, servers, and replication appliances. All Coraid products communicate using the lightweight AoE (ATA over Ethernet) protocol and standard Ethernet switches, which provides secure storage networking for industry standard x86 servers.

Coraid EtherDrive SAN promises an impressive list of capabilities, including:

• Price-performance: Higher performance than comparable Fibre Channel configurations, at approximately 20% of the cost.
• Massive throughput: More than 1200 MB/sec of throughput per Coraid EtherDrive SRX-Series storage array shelf for large-block sequential workloads.
• Simple scalability: Ease of implementation and management of Coraid EtherDrive storage compared to Fibre Channel and iSCSI.
• Optimized for virtualization: VMware and Hyper-v see Coraid storage as local-attached disks, with no need for switch configuration or multi-pathing software.

ESG Lab’s testing was designed to explore Coraid’s EtherDrive SAN and the AoE protocol, paying special attention to ease of use and management, capacity and performance scalability, and integration and operation in virtualized environments.

ESG Lab Validation

ESG Lab performed hands-on evaluation and testing of Coraid’s EtherDrive SAN at Coraid’s Redwood Shores, CA headquarters. Testing was designed to demonstrate the ease of installing and configuring an EtherDrive SAN as well as the cost-effective performance and capacity scalability of the platform.

Background: Ethernet SAN

Coraid’s EtherDrive SAN utilizes the AoE protocol to present disk storage to servers across a standard Ethernet network. AoE is an extremely simple method for sharing disk drives through a network. The communication that would normally take place between a motherboard and an IDE disk drive is arranged into data packets and sent across the Ethernet.  As can be seen in Figure 3, AoE is a simpler and more direct protocol than either iSCSI or Fibre Channel. AoE is not built on IP, TCP, or SCSI; packets are addressed to devices using their Ethernet MAC addresses and sent across the network with a minimum of overhead.

Fibre Channel and iSCSI are both based on SCSI, which is a complex protocol designed for a variety of devices (scanners, printers, etc.), in addition to disk drives. Because of this, they incur significant overhead when processing each packet. Both Fibre Channel and iSCSI run SCSI over high level networking protocols on top of a physical network infrastructure, consuming additional overhead and processing compared to AoE, which connects servers and storage directly across the physical Ethernet layer. The typical AoE packet contains just 48 bytes, plus the data payload, enabling “bare metal” performance and native Layer 2 multi-pathing. Fibre Channel and iSCSI first encapsulate the data in the SCSI command set and then wrap SCSI in a transport protocol.

Because they do not run over high level networking protocols like IP, AoE packets (like Fibre Channel) are non-routable. While they can travel across the switches that make up an Ethernet LAN, routers cannot send them to another network and devices outside of the AOE devices local network cannot communicate with them.  This makes AoE packets intrinsically secure. Coraid enables remote access to EtherDrive SANs for administration via AoE tunneling, which is similar to VPN access to a corporate network over the internet.

Getting Started

ESG Lab testing was conducted on a pre-wired, rack-mounted environment consisting of multiple SR2421 and SRX3500 EtherDrive SAN disk shelves. The ESG Lab test bed, as presented in Figure 4, consisted of multiple industry-standard x86 servers with both 20Gbps Coraid HBAs and 1Gbps Ethernet NICs installed. Servers were running VMware ESX server with Red Hat Linux and Windows 2008 installed as guest operating systems as well as physical Linux and Windows 2008 installations. An industry standard Ethernet switch was used for SAN connectivity.[3]

ESG Lab Testing

ESG Lab testing began by powering on an SRX3500 EtherDrive SAN shelf, then logging into a Linux server. Coraid’s cec utility was used to scan for the new chassis using the AoE protocol. In less than a minute, the shelf was visible.

The next step was to name the shelf to make it easier to identify it in a large deployment. Shelf 3 was chosen as the name for these tests. Next, using just three commands, RAID groups were created (Coraid automatically creates one LUN per RAID group), hot spares were assigned, and the LUNs were brought online, as seen in Figure 5.

LUN masking, the means by which servers are given exclusive access to volumes in a SAN environment, is done by Ethernet MAC address using the “mask” command. ESG Lab did not use LUN masking in these tests.

On the Linux server, ls /dev/etherd showed all AoE devices on the network. The storage administrator has nothing else to do—no iSCSI mount, no NFS mount.  The AoE LUNs look like local storage. Next, ESG Lab used mkfs to create and format a file system on each of the AoE LUNs.

Creating LUNs and presenting them for use on the network took less than one minute, while creating the file systems for use by the server took about another minute. In less than two minutes and just four simple commands, ESG Lab configured, provisioned, and was using Coraid EtherDrive storage.

Disruptive Price-Performance

Coraid EtherDrive SAN storage is a modular disk storage system providing massive scale-out capacity and performance with granular, just-in-time scalability to industry standard, open systems environments. The Coraid solution scales by simply installing additional disks and shelves, allowing organizations to start small and scale capacity to petabytes. Using 2 TB SATA drives, users can scale to a petabyte of capacity and 100 GB/sec of raw storage bandwidth in just two racks.

Performance in a storage environment is best measured with the metrics used by the applications organizations actually run.  For an e-mail application, that measurement is the number of users or mailboxes a given system can support.  For a streaming media application, the number of objects served concurrently that can be sustained during peak periods of activity is the measurement that matters most.

ESG Lab Testing

Performance was tested using the IOMETER workload generator via simulated application workloads based on Microsoft Exchange and streaming media services.  Tests were performed to verify a Coraid platform’s ability to deliver predictably scalable performance in a clustered scale-out environment over a standard Ethernet network. The Exchange workload is random in nature and very disk intensive.

Microsoft guidelines recommend a maximum of 1,000 Exchange users per core and less for a server performing multiple roles. This means that a quad-core server, doing nothing but Exchange, should support about 4,000 users.

Microsoft’s IOPS per mailbox guidance for Exchange 2007 is calculated based on the number of messages per mailbox, the user memory profile, in what Outlook mode the mailboxes are operating, and whether any third party mobile devices are used. The baseline value provided by Microsoft is .32 IOPS per mailbox.[4] This means that a quad core Exchange server with 4,000 exchange users will, on average, drive 1,280 IOPS to the Exchange Datastore.  As can be seen in Figure 6, a single SRX3500 LUN was able to support enough transactional IO to support more than 4,500 Exchange users using just 12 SAS drives and scaled linearly to just over 9,000 users with 24 SAS drives.

Next, streaming media performance was examined. This type of traffic is sequential in nature and uses larger block sizes than transactional workloads, putting more of a load on the storage network.

As Figure 7 shows, streaming media performance was excellent, delivering 826 MB/sec from just 6 SSD drives and more than 1,200 MB/sec from 24 SATA drives. Put into perspective, a single shelf was able to drive enough bandwidth to saturate a 10Gbps interface.

The maximum throughput recorded (1200+ MB/sec) was used to calculate the number of streams that could be delivered for a couple of well-known content types including standard definition and high definition broadcast video.  Bit stream rates of 3.75 Mbps for standard definition broadcast video and 80 Mbps for high definition video were used to determine that a single SRX3500 has the bandwidth required to simultaneously stream 120 high definition broadcast videos or 2,560 standard definition broadcast videos as shown in Figure 8.

What the Numbers Mean

• The system showed excellent disk response times for both random and sequential IO. The simulated Exchange disk IO response time was 20ms, while streaming media requests from SATA disk were satisfied in just 1ms.
• Microsoft stresses that, to ensure a positive user experience, the Exchange database LUN requires read and write response times of 20 milliseconds or less so that Exchange can service users’ client software quickly and efficiently. In this context, the SRX 3500’s performance is right on target.
• A single SRX3500 has the raw bandwidth required to service 2,560 concurrent standard definition, broadcast-quality video streams.

Next, ESG lab examined cost of acquisition for a petabyte of storage and SAN connectivity for various technologies. Each storage technology was configured to support the same class and quantities of storage, and SAN connectivity was calculated to support 200 physical servers with redundant connections. Table 2 summarizes the configuration built for each technology.

The cost of storage and SAN connectivity hardware was obtained from a combination of publically available sources, including reseller websites, GSA pricing schedules, and online pricing available directly from vendors.

The cost was calculated for modular dual controller Fibre Channel SAN arrays from three major vendors. The cost of dual controller multi-protocol arrays from two major vendors and the cost of direct attached storage (DAS) solutions from two major vendors were also calculated. The solution with the lowest overall price in each category was used for the comparisons presented in this report.

The bottom line results are summarized in Figure 9. Note that the costs of iSCSI, multi-protocol, and FC SAN solutions are significantly higher than a comparable Coraid EtherDrive SAN system and that the base costs of a Coraid SAN solution are lower even than DAS.

Calculated costs are detailed in Table 3.

What the Numbers Mean

• Coraid EtherDrive SAN has the lowest cost of acquisition, by a wide margin.
• The relative cost of acquisition of alternative technologies ranges from roughly 1.4x for DAS to more than 5x for FC SAN.
• The FC SAN solution is so much more expensive in part due to the cost of acquiring FC SAN connectivity.
• DAS technology has a number of limitations that were not considered in this analysis. First and foremost, it is a dead-end when it comes to server virtualization. SAN attached storage is needed to take full advantage of the benefits of server virtualization. Storage capacity held captive within, or directly attached to, a server can’t be moved non-disruptively to another server for maintenance or better quality of service. SAN attached storage is also needed to achieve valuable disaster recovery capabilities that have recently become available from server virtualization vendors (e.g., VMware Site Recovery Manager). And finally, islands of DAS capacity typically lead to poor storage utilization. Poor storage utilization dramatically increases the overall cost of ownership.
• In addition to CAPEX, ESG Lab believes it is likely that Coraid EtherDrive’s simplified architecture and management would also yield OPEX savings over alternate technologies.

ESG Lab also compared price-performance for the Coraid EtherDrive SAN systems tested to publically available results published for DAS and traditional Fibre Channel SAN systems. Price-performance was determined using a simple calculation of cost in dollars for a specific configuration divided by the number of MB/sec supported by that platform.

Virtualization Optimized

Coraid EtherDrive SAN storage systems integrate with VMware using a simple driver that enables VMware to mount EtherDrive storage arrays as if they were local drives. A VMware administrator can provision and manage virtual machine storage without the need for FC SAN administration or iSCSI client configuration.

ESG Lab Testing

ESG Lab performed virtualization tests on a VMware ESX 4.0 environment with two physical servers and six virtual machines.

First, ESG Lab logged into the vSphere client and clicked on server 192.168.0.214. As seen in Figure 10, the Coraid EtherDrive HBA was visible in the list of storage adapters and volume 10, created using the steps in Figure 5, was visible and ready for use.

The volume was formatted and made available to virtual machines using the Add Storage wizard, shown in Figure 11.

Next, the volume was assigned to a virtual machine using the native VMware Add Hardware wizard. Once the addition was complete, the volume was visible to the Windows operating system on the virtual machine. Figure 12 shows the Windows Disk Administrator tool with the new drive circled in green.

Finally, ESG lab examined availability, testing the synchronous mirroring capability of the Coraid EMX EtherDrive Mirror Appliance as well as the ability to physically move disk drives between chassis without disruption.

The availability test bed, depicted in Figure 13, consisted of three Coraid EtherDrive SR2421 shelves, one EMX Mirror Appliance, and one vSphere server, with one virtual machine running Windows Server 2008.

Two 12-disk RAID5 LUNs were created on two separate shelves and synchronously mirrored through the EMX appliance. Mirroring two volumes using the EMX appliance could not have been simpler. The mkmir command was used to select the source and target volumes to be mirrored. This single command pairs the volumes and starts the synchronization.

Next, the volume was assigned to a Windows server 2008 VM on the vSphere server. Once the volumes were fully synchronized, an IOmeter workload was started on the server, performing a mixed read/write workload against the volume, set to continue indefinitely. Power to the primary SR shelf hosting one side of the mirror was killed. Iometer continued reading and writing to the volume with no errors.

Finally, a single eight-disk RAID5 LUN in a single chassis was used to test the online drive relocation capability of the Coraid architecture. The LUN was assigned to a Windows 2008 VM and an IOmeter workload was started on the server, again performing a mixed read/write workload against the volume, set to continue indefinitely.

Power was killed to the chassis housing the eight-drive RAID 5 LUN. All eight disks were then physically relocated from the primary chassis to a spare chassis. The spare chassis was then renamed to have the same shelf number as the original chassis and the eight-disk LUN was placed online.

Total time for this physical failover was approximately three minutes. After the LUN was placed back online, the IOMeter transactions resumed successfully with no further service interruption. Most, if not all, other architectures, including highly available Fibre Channel and iSCSI SANs, simply cannot take LUNs offline in a VMware environment while machines are running without bringing the server to a crashing halt.

ESG Lab Validation Highlights

• ESG Lab configured, provisioned, and was utilizing Coraid storage in less than two minutes from power on.
• The SRX3500 demonstrated the ability to support thousands of Exchange users using just 12 SAS drives.
• Coraid EtherDrive SAN was able to drive more than 1200MB/sec from a single appliance, enough to stream 2,560 broadcast quality video streams simultaneously.
• Commodity hardware and cost-efficient AoE connectivity enable a cost of acquisition far less than Fibre Channel, iSCSI, and even DAS.
• Coraid proved well-suited to virtualized environments, providing simple to provision SAN storage that looks to a VMware cluster like direct attached disk.
• The EMX Mirroring appliance provided synchronous data protection for volumes across shelves with no disruption to service.
• ESG Lab was able to remove drives that were actively being accessed and move them to a different chassis with only a momentary pause in IO and no errors.

Issues to Consider

• Coraid’s EtherDrive SAN is currently managed through a command line with no GUI. The system is incredibly simple to use and manage, with all necessary functions controlled through a few simple commands and logical, human readable addressing of shelves, disks, and LUNs. Coraid indicated plans to ship an upgraded management system in Q3 2010 with a GUI and REST API support.
• The Coraid EtherDrive SAN solution does not yet offer advanced storage virtualization functionality such as thin provisioning or storage tiering. The driving factor behind these features, reducing the cost of storage, does not necessarily affect Coraid as it does traditional SAN architectures, which typically sell for many multiples of Coraid’s acquisition cost. In addition, these features are increasingly available in software at the hypervisor or file system layer, further obviating the need for them as array-based features.

The Bigger Truth

With storage costs consuming at least 28% of IT budgets,[6] companies are under constant pressure to find ways to reduce costs. Taking a long hard look at reducing capital and operational costs in the storage environment makes sense and so today, more than ever, IT is investing in new technology with a clear focus on reducing storage costs.

The high capacity and performance requirements of scale-out applications including backup to disk, content delivery, server and desktop virtualization, clustered computing, rich media, and un-structured bulk storage are taxing the budgets and infrastructure of IT organizations. Traditional storage network infrastructure can provide the capacity, agility, and performance these applications need, albeit at a high cost of entry and daunting complexity.  Rows of equipment are often needed to provide a petabyte of capacity and gigabytes per second of throughput. Data center managers are being pushed to the limit as administrators spend more and more time managing an ever-expanding SAN infrastructure.

ESG Lab found that the Coraid EtherDrive SAN storage system delivers shockingly simple deployment and management, with complete functionality delivered via a handful of easy to use commands and rock solid Ethernet SAN connectivity delivered via the extremely lightweight AoE protocol. The ease of management, deep scalability, and performance required for bandwidth-intensive scale-out applications are seamlessly extended to VMware environments as well.

ESG Lab testing has confirmed that Coraid’s architecture provides consistent levels of throughput—even during hardware faults.  Sustained throughput in excess of 1,200 MB/sec was observed for large block sequential reads. Cost-efficiency was impressive, with acquisition costs as low as 20% of the costs of traditional SAN attached storage. ESG Lab also verified a very interesting recoverability and resiliency feature, whereby drives can be moved to a spare chassis while an application is running.

With EtherDrive SAN storage, Coraid has dramatically simplified storage for consolidated and virtualized environments while enhancing performance and providing incredible cost efficiency. While the speeds and feeds are impressive, ESG Lab is most impressed by the shocking simplicity of both the AoE protocol and the Coraid architecture, making management of petabytes a reasonable task. If your organization is struggling to keep up with exponential data growth while providing ever higher levels of performance and availability, ESG Lab recommends that you consider Coraid EtherDrive SAN storage as the foundation for your virtualized data center.

Appendix

[2] Source: ESG Research Report, 2010 IT Spending Intentions Survey, January 2010.

[3] Configuration details can be found in the appendix.

[4] http://msexchangeteam.com/archive/2007/01/15/432207.aspx

[5] Source: ESG Research report, The Impact of Server Virtualization on Storage, December 2007.

[6] Source: ESG Research Report, Enterprise Storage Survey, November 2008.

Background

ESG recently completed a survey of IT professionals with a goal of understanding the growing interest in virtual desktop infrastructure (VDI).[1] As seen in Figure 1, simplification tops the list of factors driving the adoption of VDI technology. Specifically, administrators are looking to simplify the repetitive, hands-on tasks of OS and application deployments, upgrades, patch management and provisioning.  Given the budgeting and manpower challenges being driven by world-wide economic concerns, it’s not surprising that more than half of respondents indicated that that reducing capital and operational expenses is driving an interest in VDI adoption.

In order to address these challenges, a VDI solution must be easy to deploy and manage, highly virtualized, highly available, and predictably scalable. N-way clustered storage architecture is ideally suited to address all of these issues. Clustered storage aggregates multiple storage controllers into a single storage cluster. Though these clusters may contain many storage controllers, they still appear to the applications and users as a single logical pool for easy management. In traditional dual node storage systems with fixed architectures, when a user’s environment outgrows their storage system, they may be forced to buy another system to achieve greater performance or capacity. Clustered storage systems allow users to add CPU, memory, and bandwidth transparently, enabling them to scale based on the needs of the business without purchasing a whole new storage system. Such clustered architectures allow for the aggregation and virtualization of all hardware resources, performance, and capacity in a linear fashion—just-in-time and as needed.

Virtual Desktop Infrastructure with Citrix XenDesktop

Citrix XenDesktop is a desktop virtualization system that centralizes and delivers desktops as a service (DaaS) to users—anywhere. The XenDesktop solution includes virtualization software for hosting desktops, user and session management, provisioning tools, and application delivery as well as service monitoring, reporting, and support.  Leveraging the Citrix HDX suite of technologies, the solution works to optimize the user experience for diverse user scenarios. XenDesktop supports multiple forms of hosted virtual desktops, including virtual machine (VM)-based and blade workstation based virtual desktops. XenDesktop also offers a desktop streaming feature that delivers a master desktop image directly to a physical endpoint.

The HP P4000 provides a true clustered architecture with advanced features and functions that make it a compelling solution for VDI. While DAS (direct attached storage) could be used for VDI, networked HP storage provides universal access while enabling high availability, desktop mobility, and online scalability that is impossible to achieve with DAS. Figure 2 illustrates a virtual desktop environment utilizing Citrix XenDesktop and HP P4000 SAN.

Users connect to the Desktop Delivery Controller via a secure web browser session (1). The Desktop Delivery Controller authenticates a user’s credentials against Active Directory (2), and then Citrix XenDesktop assembles the user’s virtualized desktop on demand (3), using volumes residing on the HP cluster. Virtual desktop delivery is optimized via Citrix HDX technologies for low bandwidth and high latency WAN connections (4). The user has access to their personalized desktop, applications, and resources from anywhere while still benefiting from centralized desktop management in the data center.

The HP Reference Configuration for Citrix XenDesktop on XenServer

HP has collaborated with Citrix to develop a reference configuration designed to provide robust, predictably scalable VDI solutions that are jointly tested and certified. HP and Citrix have developed a number of recommended configurations and services for Citrix applications; this report focuses on Citrix XenDesktop on XenServer.

Based on HP ProLiant BL460c G6 server blades and HP StorageWorks P4500 G2 SAN storage, the reference configuration shown in Figure 3 is intended to help end-users understand the components and interactions of a VDI architecture and provide a reference end-users could use as a starting point to build a virtual desktop solution capable of supporting approximately 1,000 Microsoft Office 2007 users on Microsoft Windows XP.

Specific configurations will vary based on users’ particular environments and needs. HP suggests users work with their HP Reseller or Sales Representative to help determine the best solution for individual environments.  Also, this testing focused solely on the virtual desktop (sometimes referred to as VDI) capability within XenDesktop; the hosted, shared desktop model of desktop virtualization enabled by the XenApp capability within XenDesktop has been tested and modeled separately by HP, and reference configurations have been developed specific to that operating mode.[2]

The HP StorageWorks P4000 SAN

The HP P4000 SAN is a clustered storage system that scales to meet the needs of VDI environments with ease. HP P4000 SANs are built on enterprise-class, industry-standard platforms configured as fully contained storage nodes that provide CPU, memory, bandwidth, and capacity. Each storage node is powered by SAN/iQ storage software, which provides intelligent storage system functionality. Customers can scale performance and capacity online as needed by adding additional storage nodes without disruption to the SAN, VM’s or physical server applications. The HP P4000 SAN remains a single logical system regardless of how many storage nodes are added to it, making it just as easy to manage a 16-node cluster as it is to manage a 2-node cluster. Additionally, adding nodes to the cluster is a transparent and non-disruptive process. HP advises ESG that the average cluster size sold is 4-6 nodes. The average cluster size deployed in the field is 15-20 TB (4 nodes) and 20% of the clusters deployed in production contain more than 50TB (10 nodes).

At the core of the HP P4000 SAN’s value is  the HP SAN/iQ storage software platform, which provides SAN management features such as  storage clustering,  application integrated snapshots, thin provisioning, remote copy (asynchronous replication), and SmartClone volumes.  In addition, SAN/iQ includes the unique Network RAID feature, which protects against component and environmental failures while keeping data volumes online and accessible.  The Network RAID feature provides a level of high availability usually found only in the most expensive SAN arrays, often as an optional software component. Network RAID is included with every P4000 SAN and can be enabled, modified, or disabled online. The ability to keep a volume online and accessible is a key benefit to the VDI environment as the loss of volume access could affect dozens, if not hundreds, of desktop users.  The P4000 comes with all management functionality built-in. There is no additional software to purchase.

HP BladeSystem Enclosures and Servers

HP BladeSystem is a converged infrastructure solution for data centers of all sizes. HP BladeSystem enclosures and servers minimize energy, cooling, and space requirements by consolidating powerful physical servers into a dense chassis, while simplifying administration through IO virtualization.

The enclosure selected for the reference configuration is the HP BladeSystem c7000. The c7000 enclosure consolidates the essential elements of a data center – power, cooling, management, connectivity, redundancy, and security, in a high-density, modular package.

The server blade selected for the Enterprise reference configuration is the HP ProLiant BL460c G6, which provides enterprise-class features for high performance and reliability along with energy efficiency.

ESG Lab’s testing was designed to validate the business value of deploying an HP P4000 SAN to support a Citrix XenDesktop VDI, including capacity, performance, and operational efficiencies uniquely enabled by the HP solution.

ESG Lab Validation

ESG Lab audited the HP reference configuration for Citrix XenDesktop on XenServer and conducted hands on testing of the HP P4000 SAN with Citrix XenDesktop VDI at a Hewlett-Packard facility in Houston, Texas.

Getting Started

The test bed, shown in Figure 4, consisted of a pre-installed, pre-configured four-node HP P4000 SAN supporting a four-server Citrix XenDesktop virtual desktop environment. Two SANs were configured using HP ProCurve switches. A Windows workstation running Internet Explorer was used as a virtual desktop endpoint.[3]

XenDesktop offers both “assigned” and “pooled” hosted virtual desktops. An assigned virtual desktop provides each user with a dedicated virtual machine. Users connect to the same machine each time and all changes and personalizations persist between sessions. The assignment can either be pre-determined by the administrator or pulled from a group of available desktops and assigned on first access.

Pooled desktops are a group of virtual desktops offering a standard configuration. Users are connected to any of the available desktops and when they log off, that desktop is returned to the pool. Backgrounds, bookmarks, application settings, and other personalization can be captured separately in the user’s profile. System changes, such as installed applications, are discarded and the desktop is reset to its pristine state. This ensures that the virtual desktop always is in a known good state and the next user that connects will get a “fresh” desktop configuration. ESG Lab tested using the “assigned” method for this report.

ESG Lab Testing

ESG Lab began testing with provisioning and configuration of a new virtual desktop. A new volume was created in two steps using the P4000 CMC (Centralized Management Console), seen in Figure 5 and Figure 6.

ESG Lab right clicked on the navigation tree, seen in Figure 5 and selected New Volume, which launched the new volume dialog box.

Next, a name for the volume was created and the desired capacity was entered, as shown in Figure 6.

Finally, Figure 7 shows how the new volume was assigned to the four Citrix XenDesktop servers.

Once the volume was visible to the XenServer Resource Pool, the Citrix Access Management Console was used to add the new storage (Figure 8 ) and import a previously exported virtual machine image (Figure 9).

Figure 10 shows the newly created virtual desktop.  ESG Lab booted the virtual desktop and confirmed that it was accessible from the endpoint machine. The entire process, including storage provisioning and allocation, took less than ten minutes.

Storage Efficiency

Traditionally, virtual desktop environments are built by creating volumes that will act as a remote user’s primary hard drive, holding their operating system and applications. An administrator will create the volume for the new virtual machine and either install the client OS or (more commonly) import a previously backed up image. This image is then managed as a physical desktop would be—application and OS patches must be applied to each VM individually and each image consumes as much storage as it would on a physical machine.

HP utilizes its P4000 SmartClone technology to optimize both the allocation process and the capacity consumption of virtual desktops. As shown in the top half of Figure 12, a single ‘gold image’ virtual desktop is built and then a snapshot is taken to create a base image. Thin provisioned SmartClones (volume copies) are created from the snapshot and presented to the Citrix XenDesktop server, which sees them as independent read-writable volumes. These volumes already have the OS image and applications installed on them, so the installation or import step is not needed. HP’s thin provisioning technology operates on a zero-reservation principle, meaning that no storage is pre-allocated to a SmartClone and data is only drawn from the allocation pool as new data is written.  Figure 11 illustrates how SmartClones would be leveraged to reduce storage requirements in a Citrix XenDesktop VDI.

Virtual desktops are typically very light with regard to the amount of data written as a percentage of the volume capacity, resulting in a very space efficient environment. Based on ESG’s experience in the lab and HP’s experience in the field, ESG Lab is confident that 70 to 90 percent capacity efficiency can be achieved over the life of a P4000 SmartClone in a VDI environment.

ESG Lab Testing

ESG Lab evaluated the storage efficiency of the HP P4000 SAN in a Citrix VDI environment by creating multiple virtual desktops using a single source volume.

First, ESG Lab accessed the Citrix Access Management Console and identified the volume to be used to create the new virtual desktops. As shown in Figure 12, the XP SysPrep Image VM was examined and the 8 GB storage repository volume sr1 was identified.

Next, a snapshot was taken of the volume sr1 and as seen in Figure 13, the New SmartClone Volume wizard was invoked by right clicking on the Snapshot.

Figure 14 shows the New SmartClone Volumes dialog box. Three SmartClones were created for this test (administrators can create up to 25 at a time). Thin provisioning was specified and permission was set to read-write. Finally, the SmartClones were assigned to the XenDesktop servers, as previously shown in Figure 7.

Before the SmartClone volumes could be added to the Citrix XenDesktop environment, the source volume sr1 needed to be detached and temporarily ‘forgotten’ by XenDesktop. This allows the new volumes to be added and modified so that their UUID (Universally Unique Identifier) and VG (Volume Group) could be changed to a unique value for each SmartClone. As seen in Figure 15, the process for adding the SmartClones is exactly the same as adding ordinary storage.

Finally, ESG Lab created a new Virtual Machine, pointed it to the SmartClone, and booted it up, as seen in Figure 16.

Storage utilization was confirmed using the P4000 CMC. The three cloned desktops, with a combined virtual capacity of 24 GB, consumed less than 100 MB of physical storage in addition to the 8GB of the source volume.

Citrix Essentials StorageLink for HP P4000 SANs

The Citrix Essentials product offering includes StorageLink API integration with HP SANs to enhance the scalability and agility of both Citrix XenServer and Microsoft Hyper-V virtualization environments; enabling simplified storage set-up and operation, VM lifecycle management, dynamic server provisioning and automated site recovery for DR sites. Leveraging StorageLink technologies, Citrix Essentials integrates storage management functions into the management console for the virtual infrastructure via wizards, enabling users to utilize the advanced services native within HP storage arrays from an easy to use GUI.

ESG Lab Testing

ESG Lab walked through the process of creating a template and cloning virtual machines using Citrix StorageLink manager. As Figure 17 shows, a storage profile was created first. A storage repository was selected which contained the volume to be used as the gold image for cloning.

Next, a virtual machine template was created using the just-created storage profile, as seen in Figure 18 and a predefined hardware profile for a power user’s desktop.  The template was named SL-XPSP2-A VM Template.

Finally, the template was cloned using the Create Virtual Machines wizard in the Citrix StorageLink Manager. The template created in the previous step was selected and finally, ESG Lab entered a name, selected the hypervisor host, selected clone as the copy type, and specified 10 clones, shown in Figure 19.

The cloning process took less than three minutes and when complete, all ten virtual machines were visible in XenCenter and ready for use.

Again, storage utilization was confirmed using the HP CMC. The ten cloned desktops, with a combined virtual capacity of 100 GB, consumed less than 100 MB of physical storage in addition to the 10GB of the source volume.

Performance and Scalability

In a virtual desktop environment, performance and scalability are determined more by the number and configuration of virtual desktop infrastructure servers than by any other factor. Storage performance requirements are less predictable than traditional IT applications and a storage solution in a VDI environment must be able to meet not only the average IO requirements, but the maximum load that will be generated—typically at the start of a shift when many users will all be logging on at once—while scaling to meet the capacity needs of a large user community.

ESG Lab Testing

ESG Lab used the Iometer workload characterization tool to simulate the type of IO generated by typical desktop operating systems and applications and audited scalability testing performed by HP.[5] ESG testing was focused solely on the storage back end and performed against two, four, and 12-node HP P4000 clusters. HP tested using a more real world, holistic approach that used a live XenDesktop environment, large numbers of vDesktops, and real world applications.

HP measured multiple characteristics during testing, including IOPS, server CPU utilization and user response time to determine the scalability of the various components of the XenDesktop environment.

HP observed a range of 15 to 20 Read IOPS during vDesktop creation and a range of 5-15 mixed read and write IOPS per vDesktop during VM Boot up.  After boot up, the sustained workload observed was 2.5 IOPS, consisting of nearly all writes. The Microsoft perfmon utility was used by ESG Lab to monitor the disk traffic for a physical Microsoft XP based knowledge worker’s desktop. An average of 20 IOPS was observed over multiple eight hour business days. With these data points in mind, a conservative value of 20 IOPS per virtual desktop user was used to estimate the number of virtual desktops that can be supported by an HP XenDesktop infrastructure.

Results recorded by ESG using the Iometer workload characterization utility, as well as results measured and projected by HP are detailed in Table 1. HP tested with one host running XenDesktop on XenServer against one two node P4000 cluster. Projections for larger configurations were based on the amount of IO that one server could drive. ESG tested with multiple virtual machines running Iometer, and were designed to determine the maximum amount of ‘desktop-like’ IO that a cluster could support. The results are summarized graphically in Figure 21.

ESG observed that the HP P4000 storage cluster scales IO nearly linearly. HP’s testing and projections for a real XenDesktop environment tracked closely to ESG’s testing. When sizing a complete VDI solution, care must be taken to follow vendors’ best practices for server sizing and configuration as well as storage.

What the Numbers Mean

• The number of virtual desktops that the infrastructure can support scales nearly linearly as servers and storage nodes are added to the HP P4000 cluster.
• HP’s reference configuration performance projections match closely with the observed performance results of ESG Lab HP P4000 testing.

High Availability

The HP P4000 architecture addresses availability at multiple levels. Hardware-based RAID technology is used within each server in a SAN/iQ storage cluster as a first line of defense against hard drive failures. In addition, SAN/iQ stripes data across all of the nodes in a cluster. In addition the Network RAID feature provides the option of spreading one or two extra copies of data throughout the cluster to protect against data loss due to the failure (or loss of connectivity) of a server participating in the cluster. A “stretched cluster” approach is also supported. For example, one half of a cluster could be located in a data center and the other half in a second location on a campus or in a building. In this manner, data loss can be avoided due to a localized facility error that affects half the nodes in the cluster (e.g., an overloaded power circuit or network failure).

ESG Lab Testing

Availability testing was performed against a stretched four node cluster on two separate gigabit Ethernet networks connected with a 10 gigabit Ethernet uplink. The Citrix XenDesktop servers had connectivity to all nodes in the stretched cluster.

ESG Lab introduced a variety of errors to validate fault tolerance. The stretched four-node SAN/IQ cluster shown in Figure 22 was used for hardware error injection testing. The following errors were injected as an Iometer workload was being run continuously on a virtual desktop running in the Citrix XenDesktop cluster as seen in Figure 23:

• Pulled a back-end Ethernet interface on node 1 at  Site 1
• Pulled an active disk drive
• Replaced the pulled drive
• Removed connectivity to both nodes at Site 1 from the cluster

Through all injected faults, Iometer continued to run on the virtual desktop without interruption. Next, ESG Lab transitioned the running virtual desktop between Citrix XenServers using Live Migration. Disk IO paused for a few seconds while the transition occurred, but the endpoint never lost connectivity and Iometer continued to run without error.

Finally, ESG Lab simulated a site failure by downing the Citrix XenDesktop server at Site 1. All VMs automatically transitioned to the other running XenDesktop servers in the cluster. Connectivity was lost to the running virtual desktop, but our simulated user was able to immediately log back in and bring up their desktop.

ESG Lab Validation Highlights

• ESG Lab found the HP P4000 SAN easy to configure, implement, and manage in combination with the Citrix XenDesktop environment.
• ESG Lab was able to use one ‘gold image’ virtual desktop to create and present multiple unique virtual desktops with minimal capacity overhead and no impact to users.
• The efficiency and cost effective scalability of the HP architecture was seen to meet the performance needs of real-world applications deployed in a distributed virtual desktop environment.
• The HP P4000 SAN was able to sustain continuous access for a Citrix XenDesktop user through disk, network, node, and site failures and a virtual desktop was able to automatically transition to a running XenDesktop server after a simulated server failure.

Issues to Consider

• While ESG tested back-end storage performance in a virtual desktop environment, other factors, including the CPU and memory configuration of the infrastructure servers and virtual machines, will have a much greater impact on the end-user experience. ESG Lab recommends that end-users leverage reference configurations and work with their virtual infrastructure vendor to determine the best practices and optimal configuration for each environment.
• While leveraging the P4000’s Snapshot and SmartClone technology for virtual desktop deployment and management is compelling and powerful, the process of making the cloned desktops unique after importing into XenDesktop was manual when ESG Lab first tested for this report. Subsequent demonstrations with HP confirmed that Citrix Essentials for XenServer now includes StorageLink Connect API integration with HP SANs to automate the process and make it more user-friendly. Leveraging StorageLink technologies, Essentials for XenServer integrates storage management functions into the management console for the virtual infrastructure via wizards enabling users to utilize the advanced services native within HP storage arrays.

ESG Lab’s View

Increasing numbers of clients and applications make desktop management a daunting task for IT. The number of applications supported increases with organization size, compounding desktop management challenges for large organizations. With increasing numbers of corporate applications to support, ongoing maintenance and management tasks directly translate into considerable IT staffing requirements and costs. Like server virtualization, desktop virtualization is establishing a foothold in the data center among IT staffs looking to optimize their current PC environments.

HP’s P4000 SAN has a highly scalable, clustered architecture that simplifies management and allows customers to start at the level of capacity and performance they require today and grow their environments on demand. Additionally, it is easy to use and manage, while providing advanced features such as Network RAID, Smart Clones, remote replication, and thin provisioning.

Customers can stretch their clusters to create multi-site SANs. We have seen storage systems that scale in this fashion with NAS and CAS products, but in our opinion, HP is one of the leaders in SAN attached true N-way clustered storage. ESG has long been a proponent of scalable clustered storage and we believe it will become the dominant approach due to the compelling value it brings.

ESG Lab was very impressed with HP’s performance, as it scaled nearly linearly with a challenging small-block random IO workload and latency actually decreased as the cluster and IO load grew. ESG Lab found that HP performed well in a virtual desktop environment, providing easy provisioning and powerful integration of Snapshot and SmartClone technology to optimize capacity utilization. High availability functionality was also impressive, sustaining multiple failures while providing continuous access to attached virtual desktop users.

HP’s storage systems powered by HP SAN/iQ Software delivered an easy-to-use, flexible, scalable, highly available, and highly efficient storage solution for Citrix XenDesktop customers. Matching in storage what Citrix provides for desktops, HP SAN/iQ Software supports volume cloning for creating large numbers of virtual desktops without the delay and cost of consuming actual storage for each clone. And because it is distributed by design, creating a disaster-resilient storage infrastructure is as easy as choosing which storage modules to configure in each separate location.

HP’s close collaboration and extensive testing with Citrix ensures that the HP XenDesktop reference architectures are jointly tested, certified, and tuned to deliver optimal availability and performance, while addressing desktop management challenges in the enterprise. Through hands-on testing, ESG Lab confirmed that HP provides a robust networked storage foundation with simple configuration, powerful desktop mobility, enterprise class availability, and near-linear scalability. The HP SAN enhances the intelligence and value of Virtual Desktop Infrastructure.

Appendix

[2] www.hp.com/solutions/activeanswers/xenserver

[3] Configuration details are listed in the Appendix.

[4] Source: ESG Research Report, Virtual Desktop Infrastructure Market Trends, February 2009.

[5] Test configuration and workload details can be found in the Appendix.

[6] Source: ESG Research Report, Virtual Desktop Infrastructure Market Trends, February 2009.

[7] Source: ESG Research Report, Virtual Desktop Infrastructure Market Trends, February 2009.

Background

While deduplication can reduce the cost of the raw storage required to store and replicate backup data on disk, integration with the existing ecosystem is crucial. As shown in Figure 1, recently completed ESG research indicates that ease of implementation, performance impacts, and integration with existing backup processes are key concerns.[1] Robust management, edge to core replication, tape integration, and deduplication options are important considerations as well, especially within large enterprise-class organizations. The diverse family of Hewlett Packard data protection solutions is ideally suited to address these, and other, concerns.

Deduplication

Choosing a deduplication strategy should include a discussion on how and where deduplication should occur. There are two basic locations deduplication can occur: at the source, typically accomplished via a software agent runing on the client machine, or at the target, which involves either writing directly to the device  or running a software agent on the media server to perform deduplication.  All deduplication includes some level of overhead. If it occurs at the source, that overhead occurs on the client machine or media server and may have an impact on backup performance due to the software required on client systems, which can consume processing, storage, and/or network resources to deduplicate data.  When deduplication is perfomed at the target, the overhead is incurred in the device where data is being written.

In addition, it is important to differentiate between a pure software approach — in which the software  runs on an industry standard platform, — and an appliance-based (or storage hardware-based) solution. With a  software solution users have flexibility in their choice of physical storage, but must ensure that the system has enough I/O bandwidth to support their performance needs, sufficient system resources to support desired de-duplication rates, and the right storage security to prevent potential data loss. An appliance-based  solution has the ability to address many of these concerns, but locks users into a particular hardware platform.

HP’s D2D and VLS are target-side deduplication appliances which are designed to provide cost-effective, easy-to-deploy deduplication and are tuned for optimal performance on each hardware platform.

HP Data Protection Solutions

Hewlett Packard offers diverse data protection hardware and software to address the concerns of enterprises from the core data center to the smallest remote office. Figure 2 shows the HP family of data protection solutions as they might be deployed in a typical distributed enterprise.

HP StorageWorks VLS virtual tape libraries provide a high performance primary backup target with optional deduplication in the FC SAN enterprise data center while HP StorageWorks D2D serves as an easy to manage data protection appliance in mid market data centers and remote offices with deduplication and WAN-efficient replication across sites. HP Data Protector software provides a solution to manage the D2D and the VLS not only in sngle-site backup environments but also in replicated solutions. While this report focuses on the benefits of a total HP solution, it’s important to note that HP products integrate well with third party products. Data Protector works with any inline de-duplication solution and the D2D and VLS can be used with third party backup software. However the end to end, single vendor solution including replication enablement and integration is a unique HP offering.

Benefits of HP StorageWorks Data Protection Solutions

VLS:

• Highly scalable capacity and performance: Up to 4800 MB/sec of throughput and 1280 TB of usable storage.
• The entire capacity of the system can be presented as a single virtual library target: The system can scale without time consuming reconfiguration and rebalancing of backup software and backup jobs.
• Accelerated deduplication: Fast hardware compression, in combination with post process deduplication, provides capacity efficiency without impact to backup windows.
• WAN efficient replication: Fast, cost-effective disaster recovery capability between data centers and remote sites.

D2D:

• Distributed capacity and performance: The D2D Backup Systems offer high performance multi-streaming backup speeds of up to 720 GB/hour.
• HP Dynamic deduplication: Enables longer term data retention on disk and WAN efficient replication.
• WAN efficient replication: Fast, cost-effective disaster recovery capability between multiple remote sites.
• Ease of use and deployment: HP’s D2D Backup Systems are designed for easy installation and deployment for mid-sized business environments—it is ready to deploy right out of the box.

As of the publication of this report, HP has refreshed the entire D2D product family and introduced a new high-end model, the D2D 4312. The D2D 4312 has more processing power and offers higher capacity than previous generation D2D models.  The new D2D product family runs a new 64bit data deduplication technology called HP StoreOnce.   HP’s vision is to port StoreOnce deduplication technology to several HP platforms – including HP Data Protector.   While ESG tested the previous generation D2D appliances, the scenarios depicted and conclusions drawn in this report still apply.

Data Protector:

• Advanced Backup to Disk – 24/7 information access and quick disaster recovery.
• Multiple Recovery Point/Recovery Time Objectives – Achieve business-driven recovery objectives.
• Manage data effectively – within existing budgets and infrastructure, even as the quantity of data grows.
• Centralized Data Protection – Protect data on distributed physical and virtual infrastructures.
• Broad Interoperability – Integrates with partner solutions including NetApp, Data Domain, IBM ProtecTIER, and supports any third party inline deduplication target appliance.

ESG Lab Validation

ESG Lab performed hands-on evaluation and testing of HP’s data protection solutions at an HP facility in Fort Collins, CO. Testing was designed to demonstrate the scalability, performance, and ease of management  of HP’s solutions from the point of view of a typical enterprise systems administrator integrating HP’s disk-based solutions into an existing tape environment.

Ease of Deployment and Integration

The test environment, shown in Figure 3, was used throughout testing. Testing began with a one-node HP StorageWorks VLS9000 array with deduplication, physically installed and powered up as it would be by HP professional services for an enterprise customer, in a data center environment with HP Data Protector software installed. Other elements typical of an enterprise environment, such as physical tape libraries and  HP StorageWorks D2D appliances, were also present in the test bed.[2]

ESG Lab Testing

ESG Lab initially logged in to the Command View VLS console by pointing Internet Explorer at the administrative IP address on the VLS9000 library and entering the administrator username and password. Next, the Create Virtual Library Wizard was launched. The Create Virtual Library Wizard walks the administrator through the steps to create a virtual tape library, asking for such details as library type,  and is shown in Figure 4.

ESG Lab selected an HP ESL E-Series to match the physical library type already installed in the test environment. Next, and more important, the virtual tape drive type and quantity, as well as the capacity and quantity of virtual tape cartridges, were set, as seen in Figure 5.

In less than two minutes, the Create Virtual Library Wizard was completed and the virtual tape library was configured and presented on the SAN.  Next, ESG Lab started the HP Data Protector Autoconfigure Wizard. The HP Data Protector Autoconfigure Wizard, as seen in Figure 6, discovers new backup target devices and prepares Data Protector to use them.

The Data Protector Autoconfigure Wizard took about two minutes to discover and add the Virtual Tape Library and its virtual tape drives to Data Protector. As Figure 7 shows, within five minutes of sitting down at the keyboard, ESG Lab was running a full backup of the first server to the HP VLS9000.

Scalability and Performance

One of the fundamental advantages of VTL backup is the ability to run many backup streams concurrently using multiple virtual tape drives. A single tape drive can only perform one backup at a time.  To get more than one backup job running at the same time, more tape drives must be added and run in parallel. A disk-based backup and recovery solution with many random access disk drives emulating many virtual tape drives can run many backup jobs simultaneously.  The random access nature of disk also provides improved performance when locating individual files to be restored.

ESG Lab Testing

ESG Lab performed backups using a single node VLS9000 as a target first and then repeated the test after upgrading the VLS to two nodes in order to examine its relative performance as storage capacity is scaled.  A full backup of mulitple servers was simulated using the HP tapeperf utility set to generate data with 2:1 compressibility.[4] The first iteration ran with two servers running 10 backup streams; the second iteration was performed with three servers running sixteen streams. Performance scaled linearly when the second node was added. The screen capture in Figure 8 shows the VLS Command View Console during the two node test, running at 1,164 MB/sec.

Figure 9 shows actual performance results obtained in ESG Lab testing projected out to a fully populated, eight-node system. Detailed results for each test run are shown in Table 1.

What the Numbers Mean

• When a node was added to the VLS, performance scaled nearly linearly with no degradation.
• An eight-node system should be able to acheive 4,656 MB/sec of sustained backup throughput, representing the ability to protect nearly 128 TB of data in an eight-hour backup window.

Remote Office Protection

The family of HP disk and tape data protection solutions, combined with HP Data Protector software, can be configured to create an automated, edge-to-core data protection topology (see Figure 10) that spans multiple sites and provides disk-to-disk-to-tape (D2D2T) functionality. HP StorageWorks D2D appliances provide WAN-efficient movement of data between sites and storage tiers while Data Protector provides the single point of management and catalog for backup data—regardless of where it resides (remote office or corporate data center), what type of media it is stored on (disk or tape), or its age (recent backup or long term archive). D2D disk-based backup and replication appliances support data deduplication to reduce the resources required to store backup images on disk and replicate backup images over a WAN.

ESG Lab Testing

ESG Lab used HP Data Protector software to configure, automate, and track the migration of backup data residing on HP D2D data protection appliances, as shown previously in Figure 3. An edge-to-core D2D2T data protection strategy was implemented using an HP D2D appliance located in a simulated remote office. Remote office backup data was replicated over a simulated WAN to an HP D2D in a corporate data center with data movement carried out by the D2D systems. The object copy capabilities of the Data Protector software were used to write a copy of the data to removable media in a Fibre Channel SAN-attached HP StorageWorks MSL tape library. First, ESG Lab logged in to the D2D web management console, shown in Figure 11.

The first backup was ‘seeded’ or replicated locally over a gigabit ethernet LAN connection. Seeding is often employed by users who have large data sets at remote offices as it allows the first bulk transfer of data to complete very quickly. The D2D device is then shipped to the central data center and from then on, updates require much less bandwidth thanks to deduplication.  The first bulk replication is illustrated in Figure 12.

Replication of the first backup transferred 53 GB of data in 10 minutes and 40 seconds over an unrestricted Gigabit Ethernet connection. Once the first full backup was completely replicated to the target D2D appliance, ESG Lab inserted the ‘Network Nightmare’ WAN simulator and restricted throughput to 2 megabits per second to simulate a nearby WAN connection between the remote office and data center.

The first incremental backup was performed and automatically replicated over the 2 Mbit/sec simulated WAN connection. Replication of the incremental backup transferred 5 GB of deduplicated data in 1 hour, 17 minutes and 49 seconds.  This means the D2D transferred at the equivalent of 11 Mbit/sec over the 2 Mbit/sec WAN connection. By comparison, the second full backup resulted in the equivalent of 22.9 Mbit/sec over the same link. The higher virtual throughput is due to the greater volume of duplicate data in a full backup.

Figure 14 shows actual and projected deduplication capacity savings over 30 days of backups on a weekly full, daily incremental backup schedule. The capacity savings over 30 days was projected at 92%—an 11.94:1 deduplication ratio.

Finally, ESG Lab copied the latest full backup to tape the copy to tape capabilities of Data Protector with the D2D, which can be used for archiving. D2D frees up space by expiring source media (virtual tapes) rather than deleting the source data (files). This ensures that data that exists in multiple backups is not deleted until all references to it are deleted.

In 20 minutes, the copy to tape was complete.

Cost-Efficient Protection

Organizations of all sizes are struggling to meet the conflicting challenges associated with macro-level global financial uncertainty and micro-level information storage growth and complexity. A growing number of IT managers are turning to virtualization and consolidation technologies to meet these challenges. With a focus on scalability, automated management using rich software tools, and capacity-efficient pricing, HP’s data protection solutions are an excellent example of solutions that are purpose-built to address these issues.

ESG Lab created a total cost of ownership (TCO) model based on a hypothetical backup environment with multiple remote offices, a major data center, and remote replication for disaster protection.  The scenario examines cost-savings associated with moving from a tape-based to a VLS and D2D-based backup and recovery strategy, although it still assumes use of tape for long-term archive of backups. Costs were broken down by category:  capital expenditures, administrative costs, tape costs, maintenance costs, power and cooling costs, and total floorspace costs.  The cumulative costs for both tape- and disk-based backups were calculated annually over a five year period.  A number of assumptions were made and included in the calculations based on what a current IT organization might have in place for equipment, WAN connectivity, backup and restore policies, and capacity and performance requirements.[6] Comparisons were made between the total cost of ownership of a traditional tape infrastructure with no replication or deduplication and an HP Data Protection environment with disk-based backup, deduplication, and replication.

Figure 17 shows the total cost a hypothetical end-user would incur over five years when comparing a traditional tape environment to a backup to disk environment with replication and deduplication. The inflection point, where the disk environment becomes less costly than the tape environment, occurs just before the end of year two.

What the Numbers Mean

• The total cost of ownership of tape alone is roughly 69% higher than HP disk-based data protection with deduplication and replication.
• Data Protector software provides significant savings due to licensing based on data stored as opposed to data protected. Deduplication reduces licensing costs.
• The tape solution is more expensive in part due to the cost of acquiring tape media and the added complexity of managing the distributed tape infrastructure.
• While eliminating daily off-siting of tapes represents significant savings, tape is still the most cost-efficient method for long term archive of backups, and most organizations will replicate deduplicated data to a remote site for copy to tape for this purpose.

ESG Lab Validation Highlights

• Within five minutes of sitting down at the keyboard, ESG Lab was running a full backup of the first server to the HP VLS9000.
• The VLS 9000 demonstrated near linear performance scalability, achieving an impressive 4 TB/hour with a two-node configuration.
• The HP D2D Backup System achieved 81% bandwidth efficiency, transferring a 5 GB incremental backup across a 2 Mbit/sec simulated WAN link in just under 1 hour and 18 minutes for an effective throughput of more than 11 Mbit/sec. Replicating a second full backup yielded an even more impressive 90% bandwidth efficiency, transferring a 54 GB incremental backup across the same 2 Mbit/sec simulated WAN link in just over 6 hours and 42 minutes, for an effective throughput of more than 22.9 Mbit/sec.
• The HP Data Protection solution suite demonstrated faster, more reliable backups and restores at a significantly lower total cost of ownership than a tape environment.

Issues to Consider

• As with all VTLs today, when a cartridge is deleted or expired in a backup application, space on the VLS is not reclaimed until the cartridge is deleted, expired, or overwritten via the VLS management application. Integration with backup applications to automatically trigger a delete or expire in the VLS when a cartridge is expired in the backup application would be a useful enhancement.
• While the VLS, D2D, and Data Protector all have easy to use management interfaces, a single “manager of managers” that integrated all three products and provided an overall view of an entire enterprise environment would be of great value to administrators.
• While ESG is confident that one or more HP StorageWorks D2D Backup Systems can be used to meet the performance needs of a mid-sized organization,  D2D systems with more capacity and horsepower could reduce cost and complexity within larger mid-sized organizations.  HP has advised ESG that the new line of D2D Backup Systems, announced in June 2010, has been designed with these considerations in mind.

The Bigger Truth

ESG Lab conducted its first hands-on testing of the Hewlett Packard’s enterprise VTL, the VLS 6000, in 2006 and then validated the VLS 9000 in 2008. Testing and discussions with end-users have confirmed that HP’s disk-based backup solutions fit seamlessly into existing backup environments while providing dramatic performance and capacity reduction benefits compared to legacy tape-based methods. The HP StorageWorks D2D, aimed at delivering deduplication and WAN efficient replication to smaller, remote offices, completes a comprehensive, enterprise wide edge-to-core data protection architecture, managed by HP Data Protector, that goes beyond disk-based backup.

During this independent lab validation, ESG Lab confirmed HP’s edge to core capabilities in support of large enterprises as well as deduplication support across disk, replication, and tape throughout the product line. HP’s comprehensive offering with capacity efficient pricing provides high performance data deduplication technology to deliver dramatic disk capacity savings while offering scalable, predictable performance.

A modest two-node VLS configuration tested by ESG Lab was able to back up at a sustained 4 TB/hour.  Based on the nearly linear scalability observed by ESG,  an eight-node system should be able to protect data at 16 TB/hour. Easy to navigate, web-based management enabled ESG Lab to manage backups for a remote office from creation, through replication, and finally to tape—all using HP Data Protector software—giving rise to an edge-to-core data protection strategy covering remote and branch offices as well as multiple data centers. Direct attach to tape capability enables enterprises to meet offsite and deep archive requirements using familiar tools and techniques while keeping tape copy traffic off the SAN.

HP Data Protector software exemplifies the company’s depth and breadth of end-to-end solutions for backup and recovery encompassing disk and tape. It should bring significant value to customers grappling with the challenges associated with cost-effective management of their data protection resources.

In essence, deduplication has become a crucial component of disk to disk backups, but when considering competing methods for implementation, customers should consider the tradeoffs and what’s best for their organization: ease of implementation, cost, and bandwidth all play an important role.

ESG Lab believes that the combination of enterprise class performance and scalability—along with comprehensive storage management software and services—provides a unique approach for optimizing data protection and recovery strategy in the enterprise. Hewlett Packard customers can now retain more data for fast and reliable restores and longer retention periods while minimizing impact on backups with accelerated deduplication. Combined with the consolidated data management provided by Data Protector, customers have a wide choice of configurations which can be used to dramatically increase the role of disk in the protection of critical data.

Appendix

[2] Detailed configuration information can be found in the Appendix.

[3] Source: ESG Research Report, Data Protection Survey, April 2010.

[4] Full testing configuration is described in detail in the Appendix.

[5] Source: ESG Research Report, Data Protection Survey, April 2010.

[6] Assumptions and parameters can be found in the Appendix.

Online Digital Archiving, the Market Backdrop

Challenges, Realities, and why Archive?

ESG estimates that organizations will retain nearly 63,000 petabytes of unstructured data in digital archives over the next three years in order to meet litigation and regulatory mandates (e.g., Federal Rules of Civil Procedure (FRCP), SOX, HIPAA, GLBA, and PCI DSS) and to support business intelligence initiatives.[1] To put this in perspective, it was only in 2009, that Eric Schmidt, Google’s CEO, estimated the total capacity of data on the internet at 5,000 petabytes. [2]

ESG research indicates that the vast majority of corporate digital assets are stored as unstructured data. Unstructured file data—which includes office documents, digital images, audio, and video files—accounted for 74% of global digital archive capacity in 2009 and is expected to constitute the bulk of digital assets for the foreseeable future.  Growing at a compound annual rate of 79%, the worldwide capacity for file-based unstructured archive data clearly dwarfs that of database and e-mail data, as shown in Figure 1. The situation cries out for organization, management, and easy access.

It’s no surprise then that the majority of organizations recognize the need for file-based archiving solutions: 65% of organizations surveyed by ESG currently have some form of formal process and/or technology in place to manage the archiving of file-based content. An additional 29% of organizations surveyed expect to deploy file archiving solutions over the next few years.  In the same survey, improved search, performance, and security were high on customers’ archive solution wish-lists. Indeed, full content index and search is the number one “must have” feature among current and prospective customers of archiving solutions.

Nexsan’s Assureon: Secure, Online Archive

Assureon is a self managing content addressable storage (CAS) platform that extends Nexsan’s family of highly efficient storage systems. Assureon is a disk-based online archive platform designed to be a simple, plug-and-play solution that enables organizations of all sizes to protect, retain, search, and retrieve unstructured digital assets. Applications producing fixed content can point to an Assureon located on the network as a shared drive and begin writing data to the device immediately. This simple process enables IT generalists in small enterprises and mid-market companies to quickly deploy the system with confidence.

Assureon is engineered with the goal of providing a secure, online archive for high value content in a high performance platform that is power-, space-, and cost-efficient. Windows 2003 and newer servers use a lightweight client that provides direct archiving to Assureon of any desired drives or folders while any platform that can mount an NFS or CIFS share can access Assureon through an integrated NAS gateway, completely transparently.

Assureon provides these benefits in a flexible, scalable package that allows customers to deploy their own self-managing archive solution or to engage a service provider to provide secure hosted archive services, as shown in Figure 3.

Hands-on: Assureon in Action

Assureon online disk storage archive provides numerous features to guarantee that files are safe from prying eyes, have not been tampered with, and are not corrupted over time. ESG Lab recently tested Assureon’s online archive capabilities in a simulated content management environment, focusing on data privacy, integrity, and longevity.

Privacy

Assureon uses the following capabilities to provide secured, private access to data, both within organizations and between multiple organizations sharing the same infrastructure.

• Data Separation and Security: Individual users, departments, or customers can store their files in their own virtual archive. Data is logically and physically separated, ensuring privacy and security.
• Access Audit Trail: Assureon establishes an unalterable audit trail for the life of an archived file. Every time a file is accessed, a record is kept of who accessed it and when it was accessed.
• Encryption: Assureon protects information privacy and security with AES256 encryption, ensuring the highest levels of file privacy and security.

To provide scalability and data separation, Assureon can be divided into separate CAS archives, providing physical data separation inside an enterprise or between customers of a cloud computing SaaS provider.  Each CAS archive can house multiple virtual archives, each consisting of a separate database and CAS object store, as seen in Figure 4.

Assureon’s support for multiple virtual archives enables Assureon to scale, have no CAS object limits, and to take advantage of Nexsan’s AutoMAID reductions in power and cooling costs for infrequently accessed virtual archives. Each classification within a virtual archive can be further divided into subclasses, such as departments, for more granular control. All CAS archives and virtual archives and data classifications can be managed from a single pane of glass as a single system.

ESG Lab Testing

ESG Lab used a standard x86 PC running Windows Server 2008 with the Assureon client installed.  Using Windows Explorer, a folder named DATA was created and its properties were accessed by right-clicking on the folder name.

As seen in Figure 5, the Assureon client adds a tab to the properties page to access and set Assureon archive policies. Administrators and users will only see the Assureon tab if they have permission to set policies.

Two folders were created: Finance and Marketing. For both folders, ESG Lab set the organization name, retention rule, and action to be performed on all files created in or copied to the folders.  Archiving can be set to real time and/or sync mode. Real time causes all files to be processed immediately upon creation and close or modification and close, while sync processes files manually or according to a schedule. For the purposes of this test, the action was set to “Replace with shortcuts” to immediately replace all archived files with shortcuts. Users have the option to leave the original files in place while still protecting the files in the archive, or set a threshold to have files replaced with shortcuts when space is needed. The Assureon Web GUI can be used to create more advanced or detailed archive policies, including setting flexible or compliant retention, setting the number of days before an archived file is replaced with a shortcut, or the specific file name pattern to match, if only certain files in a folder need to be archived.

ESG Lab tested data privacy by logging in as two different users, one from the Finance department, and one from the Marketing department, then examining their archived folders using Windows Explorer, Assureon Explorer, and the content search feature in the Assureon GUI. Only the folders and files each user had permission to see were visible.

Integrity

Assureon ensures data integrity using multiple integrated techniques and technologies.

• Fingerprint and Integrity: When data is archived, it is assigned its own unique fingerprint that stays with the file through its lifecycle. If a single bit changes, the fingerprint will change.
• Self-Auditing and Self-Healing: Assureon continually monitors files for fingerprint discrepancies protecting against tampering, viruses, corruption, accidental or deliberate deletion ,and theft. If discrepancies are discovered, Assureon notifies administrators and automatically self-heals the file.
• Independent Date and Time Stamp: Assureon’s independent time source prevents modifications to the system’s time clock ensuring the integrity of retention periods. With Assureon, tampering with the system clock or the date stamp on the file itself is prevented.

To test data integrity, a text file was created and copied into the folder created in the previous step. The copy operation completed instantly, with all the archiving and housekeeping functions, such as calculating the digital fingerprint, completing automatically and transparently. The file was examined and verified that the text was exactly as entered.  Next, the file was intentionally “corrupted” by changing the contents of only one copy. ESG Lab examined the integrity log, shown in Figure 7. Self Auditing and Self Healing after the system completed an automatic integrity audit. The integrity log shows that the changes to the object were detected, corrected, and the corrupted copy of the file was copied to a quarantine area to preserve it for investigative purposes.

ESG lab examined the contents of each version of the file using Assureon explorer, confirming that the contents of the file had been restored to the correct state.

It’s important to note that Assureon explorer enables administrators to restore shortcuts, or full files, to a production server by right-clicking on the file and selecting the appropriate action.

Longevity

Longevity is the ability to guarantee that all versions of a file will remain immutable and uncorrupted over time. Assureon uses multiple techniques to provide file longevity.

• File Availability Audit: Assureon serializes each file and continually checks to ensure all files are present in each store. If a file is missing, Assureon notifies administrators and automatically returns it to its original state.
• Automated Retention and Deletion: To satisfy regulatory compliance or corporate governance, organizations need to guarantee the retention and deletion of files. With automated integrity management and file immutability technology, Assureon protects against accidental or unauthorized file deletion and ensures files can be retained for compliant or flexible time periods as well as securely deleted when their retention period expires.
• Legal Holds: Legal holds can be placed on files to protect against file deletion when retention periods have expired. This ensures availability of files in cases of litigation or potential litigation.

ESG Lab examined multiple aspects of file longevity.  The retention rules page is shown in Figure 9. Assureon has the ability to create separate sets of retention rules for every organization created in the system. Each organization can create multiple rules within each organization to provide appropriate retention for any data type.

A retention rule was created with a retention period of seven years and data encryption enabled. Next, the audit logs were examined and it was confirmed that Assureon was running both file availability and integrity audits daily, restoring corrupted and lost files as necessary.

The Bigger Truth

ESG research has found that unstructured file data accounted for 74% of global digital archive capacity in 2009 and is expanding at a compound annual growth rate of 79%. The worldwide capacity for file-based unstructured data dwarfs that of database and e-mail data combined—and growth is accelerating. Organizations must deal with all this growth while meeting compliance and governance standards and controlling the cost of search and discovery. In this scenario especially, additional complexity is not desirable.

Nexsan’s Assureon is an excellent fit for businesses of any size that require intelligent archiving with CAS. The low cost of entry makes this solution available to a whole new set of IT departments that until now simply couldn’t afford to acquire or manage such technology. Companies can implement robust data security and protection with limited IT resources, and still utilize enhanced search capabilities to quickly produce pertinent data for internal investigation and litigation support without having to incur additional expenses from service providers culling through vast archives of data located on different media types.

Assureon delivers more business value by minimizing the power, space, and cost for given capacity and performance workloads. In addition to its powerful archiving tools, Assureon allows users to move away from a traditional tiered storage implementation and to gain the added IT and business benefits of Nexsan’s highly efficient storage.

ESG Lab was quite impressed with Assureon’s transparent and simple integration into a standard server environment, integrating archive policies, classification, search, reporting, and migration into one intuitive package.  ESG Lab testing revealed that Assureon provides excellent data privacy, integrity, and longevity in a package that is as easy for small businesses to deploy and manage as it is for large enterprises or service providers. The platform scales performance and capacity independently, leveraging Nexsan’s family of highly efficient RAID enclosures and industry standard servers.  It is clear that Nexsan is continuing to execute on its vision to provide users with a highly integrated storage infrastructure.

The company does face a couple of notable challenges. First, the CAS market has never exploded in quite the way that some imagined it would and second, the CAS market that does exist also has a very dominant player (EMC). However, like so much in life, the flip side to challenge is opportunity: EMC and others have shown that a decent market does exist and it’s all open to Assureon if its channel can find a way to the table; Assureon brings a level of simplicity and functionality that could well both expand the realistic market potential and certainly give prospective users more than a moment’s pause before leaping with the default solution. If Nexsan were a brand new player, its task might be very difficult indeed—however, it is a proven, relatively small, but high quality—provider with a base of “raving-fan” users. Nexsan already has several large medical OEMs supplying Assureon to medical and PACs archive end users.

ESG Lab found the system to be extremely easy to manage, providing policy based archiving while preserving privacy, guaranteeing data integrity, and protecting data longevity from a single, simple interface.  In ESG’s opinion, Assureon will enable organizations to reduce their costs and simplify their infrastructures while providing enterprise class archiving services at the speed of online disk.

[2] Chang, Fay Dean, Jeffrey Ghemawat, Sanjay Hsieh, Wilson C. Wallach, Deborah A. Burroughs, Mike Tushar, Chandra Fikes, Andrew and Gruber Robert E. “Bigtable: A Distributed Storage System for Structured Data”, 2006, Retrieved from http://labs.google.com/papers/bigtable.html on 10/7/2008

Introduction

The use of server virtualization to consolidate server infrastructure, reduce data center floor space, and maximize utilization of existing assets has seen phenomenal growth over the past decade, but server virtualization’s considerable success is dwarfed by the potential of desktop virtualization. Increasing  variety and numbers of client device types, the mobilization of the workforce, “always-on” expectations for corporate IT services, evolving regulatory compliance mandates, tightening security policies, and a driving need to increase operational efficiency all combine to make desktop management a daunting task for even the best IT organizations.

A growing number of organizations are using VDI (virtual desktop infrastructure) technology to reduce the cost, complexity, and risks associated with desktop management while providing a high-quality, predictable, and productive computing environment. This report documents hands-on testing of an HP StorageWorks P4000 SAN in a VMware View environment—paying special attention to ease of management, performance, storage efficiency, and availability.

Background

ESG surveyed IT professionals with a goal of understanding the growing interest in VDI.[1] As seen in Figure 1, simplification tops the list of factors driving the adoption of VDI technology. Specifically, administrators are looking to simplify the repetitive, hands-on tasks of OS and application deployments, upgrades, patch management, and provisioning while improving remote users’ computing experiences.  Given the budgeting and manpower challenges being driven by global economic concerns, it’s not surprising that more than half of respondents indicated that reducing capital and operational expenses is driving an interest in VDI.

In order to address these challenges, a VDI solution must be easy to deploy and manage, highly virtualized, highly available, and predictably scalable. N-way clustered storage architecture is ideally suited to address all of these issues. N-way clustered storage supports multiple storage controllers in a single cluster, which, though it may contain many storage controllers, appears to applications and users as a single logical system for easy management. In traditional dual controller storage systems with fixed architectures, when a user’s environment outgrows their storage system, they may be forced to buy another system to achieve greater performance or capacity. Clustered storage systems allow users to add CPU, memory, and bandwidth transparently, enabling them to scale based on the needs of the business without purchasing a whole new storage system. Such clustered architectures allow for the aggregation and virtualization of all hardware resources, performance, and capacity in a linear fashion—just-in-time and as needed.

The HP StorageWorks P4000 SAN

The HP P4000 SAN is a clustered storage system that scales to meet the needs of VDI environments with ease. HP P4000 SANs are built on enterprise-class, industry-standard platforms configured as fully contained storage nodes that provide CPU, memory, bandwidth, and capacity. Each storage node is powered by SAN/iQ storage software, which provides intelligent storage system functionality. Customers can scale performance and capacity online as needed by adding additional storage nodes without disruption to the SAN, VM’s or physical server applications. The HP P4000 SAN remains a single logical system regardless of how many storage nodes are added to it, making it just as easy to manage a 16-node cluster as it is to manage a 2-node cluster. Additionally, adding nodes to the cluster is a transparent and non-disruptive process. HP advises ESG that the average cluster size sold is 4-6 nodes. The average cluster size deployed in the field is 15-20 TB and 20% of the clusters deployed in production contain more than 10 nodes.

The economics of a clustered network storage system are compelling, with the potential to significantly reduce capital and operational costs. With HP’s P4000 SAN, the customer only has to add another storage node to increase performance and capacity—which costs far less than acquiring a whole new system—and cover its associated software, licensing, and maintenance charges. Most midrange storage systems support, at best, a dual-controller configuration, which limits scalability and flexibility.

At the core of the HP P4000 SAN’s value is  the HP SAN/iQ storage software platform, which provides SAN management features such as  storage clustering,  application integrated snapshots, thin provisioning, remote copy (asynchronous replication), and SmartClone volumes.  In addition, SAN/iQ includes the unique Network RAID feature, which protects against component and environmental failures while keeping data volumes online and accessible.  The Network RAID feature provides a level of high availability usually found only in the most expensive SAN arrays, often as an optional software component. Network RAID is included with every P4000 SAN and can be enabled, modified, or disabled online. The ability to keep a volume online and accessible is a key benefit to the VDI environment as the loss of volume access could affect dozens, if not hundreds, of desktop users.  The P4000 comes with all software functionality built-in. There is no additional software to purchase.

It is important to note that in a VDI environment, the other benefits of a P4000 highly virtualized storage cluster in combination with VMware, such as thin provisioning and Linked Clones, can be leveraged to significant effect. Thin provisioning offers a simple solution to the problem of stranded capacity—it is a storage system technology that allows users to safely allocate as much logical capacity as needed to a desktop volume while physical capacity is drawn from a common pool of storage on an as-needed basis; only when a desktop application performs writes is physical capacity drawn from the storage pool. Additionally, physical capacity can be added to the storage pool non-disruptively at any time.

All virtual desktops share common data blocks. Using the VMware Linked Clones feature along with P4000 Thin Provisioning produces a highly efficient way of storing desktop data within the SAN.  Through conversations with end-users and HP, ESG Lab has observed that a range of 3% to 15% of a desktop system volume is typically consumed by unique data. Thin provisioned Linked Clones would be 90% efficient in a VDI environment where the unique data written in each desktop is equal to 10%. A 250-desktop environment where each desktop was allocated 30 GB would normally require over 7.5 TB of usable storage. Linked Clones could reduce the storage requirement in this example to less than 1 TB, a reduction of 85%.

Virtual Desktop Infrastructure with VMware View and VMware vSphere

VMware View is a desktop virtualization system that centralizes and delivers desktops as a managed service to users—anywhere. The VMware solution includes VMware vSphere virtualization software and VMware View for hosting desktops, user and session management, provisioning tools, and application delivery as well as service monitoring, reporting, and support.  VMware desktop virtualization leverages PCoIP technology, a server-centric protocol that does the majority of graphics rendering and processing on the VMware servers, transmitting only compressed bitmaps or frames to the remote clients over the LAN or WAN. Through VMware View Manager, VMware View provides a single management tool to provision new desktops or groups of desktops and a simple interface for setting desktop policies. View Composer, based on VMware Linked Clone technology, enables the rapid creation of desktop images from a master image. When updates are implemented on the parent image, they are pushed out to any number of virtual desktops in minutes, simplifying deployment and patches without affecting user settings, data, or applications.

Figure 2 illustrates a virtual desktop environment utilizing VMware View and the HP P4000 SAN.

Users connect to the VMware View Manager using either the VMware View client or any one of numerous certified third party hardware and software clients.[2] VMware View Manager authenticates a user’s credentials and then uses those credentials to automatically authenticate users as they log into their virtual desktops, using volumes residing on the HP P4000 cluster. The virtual desktop is delivered via the PCoIP optimized delivery protocol. The user has access to their personalized desktop, applications, and resources from anywhere while still benefiting from centralized desktop management in the data center.

ESG Lab’s testing was designed to validate the business value of deploying an HP P4000 SAN to support a VMware View VDI, including capacity, performance, and operational efficiencies uniquely enabled by the HP P4000 SAN .

ESG Lab Validation

ESG Lab conducted hands-on testing of the HP P4000 Virtualization SAN with VMware View VDI at a Hewlett-Packard facility in Houston, Texas.

Reducing Complexity

The test bed, shown in Figure 3, consisted of a pre-installed, pre-configured four-node HP P4000 Virtualization SAN supporting a VMware View virtual desktop environment. Two P4000 SANs with two nodes each were configured using HP ProLiant DL385 G6 servers and HP ProCurve switches for the IP traffic between the SANs. Windows workstations running Windows XP were used as virtual desktop endpoints.[3]

VMware View offers both persistent and non-persistent hosted virtual desktops. A persistent virtual desktop provides each user with a dedicated virtual machine. Users connect to the same machine each time and all changes and personalization persists between sessions. The assignment can either be pre-determined by the administrator or pulled from a group of available desktops and assigned on first access.

Non-persistent desktops are a group of virtual desktops offering a standard configuration. Users are connected to any of the available desktops and when they log off, that desktop is returned to the pool. Backgrounds, bookmarks, application settings, and other personalization can be captured separately in the user’s profile. System changes, such as installed applications, are discarded and the desktop is reset to its pristine state. This ensures that the virtual desktop is always in a known good state and the next user that connects will get a “fresh” desktop configuration. ESG Lab tested using the persistent method for this report.

ESG Lab Testing

ESG Lab began testing with provisioning and configuration of a new virtual desktop. A new volume was created in two steps using the HP P4000 Centralized Management Console, seen in Figure 4 and Figure 5.

ESG Lab right clicked on the navigation tree, seen in Figure 4 and selected New Volume, which launched the new volume dialog box.

Next, a name for the volume (SH-L001) was created and the desired capacity (500 GB) was entered, as shown in Figure 5. Finally, Figure 6 shows how the new volume was assigned to the four VMware View servers.

Once the volume was visible to the VMware cluster, the vSphere Client management console was used to rescan the adapter for the new storage and add the volume to the server, shown in Figure 7.

Next, a virtual machine was deployed from a template to provide a master image for cloning large numbers of virtual desktops, shown in Figure 8.

ESG Lab booted the virtual desktop and confirmed that it was accessible from the endpoint machine. The entire process, including storage provisioning and allocation, took just over seven minutes.

Storage Efficiency

One way of provisioning virtual desktop environments  is to build fully provisioned volumes that will act as a remote user’s primary  computing environment,  including their operating system and applications. An administrator will create the volume for the new virtual machine and either install the client OS or (more commonly) import a previously backed up image. This image is then managed as a physical desktop would be—application and OS patches must be applied to each VM individually and each image consumes as much storage as it would on a physical machine.

HP P4000 utilizes its thin provisioning technology in combination with VMware Linked Clones to optimize both the allocation process and the capacity consumption of virtual desktops. Figure 9 compares traditional allocation and provisioning with thin provisioned Linked Clones. When using thin provisioning and Linked Clones, a virtual desktop is built and then a snapshot is taken to create a parent image of a virtual machine. Linked Clones are created from the snapshot using VMware View, which sees them as independent read-writable volumes. These volumes already have the OS image and applications installed on them, so the installation or import step is not needed. HP P4000’s thin provisioning technology operates on a zero-reservation principle, meaning that no storage is pre-allocated to a parent image or Linked Clone and data is only drawn from the allocation pool as new data is written.

Virtual desktops are typically very light in terms of the amount of data written as a percentage of the volume capacity, resulting in an environment that is an excellent candidate for thin provisioning. Based on ESG’s experience in the lab and HP’s experience in the field, ESG Lab is confident that 70% to 90% capacity efficiency can be achieved over the life of a VMware Linked Clone in a VDI environment.

ESG Lab Testing

ESG Lab evaluated the storage efficiency of the HP P4500 Virtualization SAN in a VMware VDI environment by creating 72 virtual desktops using a single parent image.

First, ESG Lab accessed the VMware vSphere client and identified the virtual machine to be used as the parent to create the new virtual desktops. As shown in Figure 10, the virtual machine master snapshot (parent image) was created and named SS-XPM-01.

Once the snapshot was created, VMware View manager was used to deploy a pool of desktops based on the parent image via a deployment wizard, as detailed in the next few screenshots. First, the wizard asks what type of desktop configuration is being deployed; ESG Lab selected “automated pool.” An automated pool contains desktops that are automatically created and customized by View Manager based on a VMware vCenter virtual machine template (the snapshot created in the earlier step).

Next, the desktops being created were specified to be persistent. Persistent virtual desktops are assigned to their user on the first use, so the user returns each time to the same virtual desktop. This type of pool is used when users want to customize their desktops by installing additional applications and storing local data. Non-persistent desktops are used in environments where desktops are only required temporarily and can be deleted after every use to give users a clean desktop every time they log in.

Figure 11 shows the next step, specifying the VMware vCenter servers that will be utilized for this desktop pool and the use of Linked Clones.

Next, a unique ID and desktop settings were specified for the pool, followed by provisioning settings, where the number of desktops was specified as 24 for this pool. Several steps followed where the parent VM was identified as well as the default image snapshot and the cluster on which to run the virtual desktops was selected as well as a location to store users’ data.

Figure 12 shows the datastore selection screen. Storage overcommit, which determines how aggressively the system assigns new VMs to the space available on a datastore, was left at the default (conservative).

After specifying the datastore, QuickPrep settings were specified, which are used to configure desktops after they have been created, including joining them to a domain, if necessary.

Figure 13 shows the summary screen of the Add Desktop wizard. When ESG Lab clicked “Finish,” the 24 virtual desktops were created and configured.

The entire process to create 24 desktops, from the first click of the wizard to the desktops being ready for use, took 35 minutes to complete. ESG Lab repeated the Add Desktop wizard twice to create a total of 72 desktops. The entire process of creating 72 new persistent desktops from start to finish took 101 minutes, or less than one and a half minutes per desktop.

Storage utilization was confirmed using the HP P4000 CMC. The 72 cloned desktops, with a combined virtual capacity of 239 GB, consumed less than 10% of the projected capacity of 2.88 TB based on the 40 GB volume size of the template VM.

Performance and Scalability

In a virtual desktop environment, performance and scalability are determined more by the number and configuration of virtual desktop infrastructure servers than by any other factor. Storage performance requirements are less predictable than traditional IT applications and a storage solution in a VDI environment must be able to meet not only the average IO requirements, but the maximum load that will be generated—typically at the start of a shift when many users will all be logging on at once—while scaling to meet the capacity needs of a large user community.

To reduce potential complexity for users, HP uses the phrase “POD” in a VDI environment to describe the aggregated resources required to host a given number of users. A POD contains any number of virtual infrastructure servers and an HP P4000 SAN system consisting of a specific number of storage nodes as well as software and thin clients. The servers provide the processing power to run additional virtual desktops while the storage nodes provide both additional capacity and storage performance to the cluster.

ESG Lab Testing

To test storage performance in a VDI environment, ESG Lab used a proprietary HP load generation tool, which captures actual desktop IO and plays it back to simulate as many desktop sessions as desired. The HP tool was run against storage clusters of three, four, and five nodes. Previously collected Iometer workload characterization results were audited for 10-, 15-, 20-, 25-, and 30-node environments. ESG Lab used the Iometer workload characterization tool to simulate the type of IO generated by typical desktop operating systems and applications.[4]

HP indicates that a broad range of IOPS has been observed for virtual desktop workers in the field.  Additional internet research reveals 5 to 6 IOPS is frequently indicated as a typical value found for Windows XP workstations, but applications and workloads run by users may alter this considerably. ESG Lab ran the Microsoft perfmon utility to monitor the disk traffic on a physical Microsoft XP desktop for a knowledge worker with heavy IO usage. An average of 20 IOPS was observed over multiple eight hour business days. With these data points in mind, a conservative value of 20 IOPs per virtual desktop user and an optimistic value of 5 IOPS per desktop were used to estimate the number of virtual desktops that can be supported as HP P4000 storage nodes are added to the HP P4000 SAN system.

IOPS results recorded by the HP tool and the Iometer workload characterization utility, detailed in Table 1, were used to estimate the number of virtual desktops that might be supported for each of the configurations.  Both the conservative estimate of 20 IOPS and the optimistic value of five IOPS per virtual desktop were used for these calculations. The results are summarized graphically in Figure 15.

What the Numbers Mean

• The number of virtual desktops that the infrastructure can support scales nearly linearly as storage nodes were added to the HP P4500 Virtualization SAN.
• As the maximum performance rose, response time got shorter due to the larger pool of storage nodes and drives which were available to respond to IO requests.

Availability and Recovery

The HP P4000 SAN architecture addresses availability at multiple levels. Disk based RAID technology is used within each storage node in a P4000 SAN as a first line of defense against hard drive failures. In addition, the P4000 cluster stripes data across all of the nodes in a cluster. The Network RAID feature provides the option, volume-by-volume, to create multiple mirrors of data throughout the cluster to protect against data loss due to the failure (or loss of connectivity) of a storage node participating in the cluster. A “stretched cluster” approach is also supported, where one half of a cluster could be located in a data center and the other half in a second location within a building or on a campus. In this manner, data loss can be avoided due to a localized facility error that affects an entire data center (e.g., an overloaded power circuit or network failure).

ESG Lab Testing

Availability testing was performed against a  six-node P4000 Multi-site SAN comprised of two separate gigabit Ethernet networks connected with a 10 gigabit Ethernet uplink, as shown in Figure 16 . The VMware vSphere servers had connectivity to all storage nodes in the stretched cluster.

In the first phase of testing, ESG Lab introduced a variety of hardware errors to validate fault tolerance. The following errors were injected as an Iometer workload was being run continuously on a virtual desktop running in the VMware View cluster as seen in Figure 17:

• Pulled a back-end Ethernet interface on node 1 at site 1.
• Pulled an active disk drive.
• Replaced the pulled drive.
• Removed connectivity to all nodes at site 1 from the cluster.

Through all injected faults, Iometer continued to run on the virtual desktop without interruption. Finally, ESG Lab simulated a site failure by removing network connectivity to all storage nodes at site 1. There was a pause of approximately 20 seconds, then user access resumed. Users were not disconnected and could continue working after the pause in IO.

ESG Lab Validation Highlights

• ESG Lab found the HP P4000 SAN easy to configure, implement, and manage in combination with the VMware View environment.
• ESG Lab was able to use one “parent image” virtual machine to create and present multiple unique virtual desktops with minimal capacity overhead and no impact to users.
• The efficiency and scalability of the HP P4000 SAN architecture was seen to meet the performance needs of real-world applications deployed in a distributed virtual desktop environment.
• The HP P4000 SAN was able to sustain continuous access for a VMware View user through disk, network, node, and site failures. Each simulated failure immediately triggered an alert that was sent to a systems administrator.

Issues to Consider

• While ESG tested back-end storage performance in a virtual desktop environment, other factors, including the CPU and memory configuration of the infrastructure servers and virtual machines, will have a much greater impact on the end-user experience. ESG Lab recommends that end-users work with their virtual infrastructure vendor to determine the best practices and optimal configuration for each environment.
• While leveraging VMware’s snapshot and Linked Clone technology for virtual desktop deployment and management is compelling and powerful, it is a value-added feature that is licensed per user in a VMware environment. A bundling option which leverages the Snapshot and SmartClone technology built into the P4000 SAN and avoids the VMware license fee would make this an even more cost-effective solution.
• It is important to note that the Iometer utility, used for some of the performance testing in his report is optimized for generating IO and does not place great demands on server memory or CPU. The performance tests conducted for this validation focused on storage scaling and did not include server sizing considerations. When sizing a complete VDI solution, care must be taken to follow vendors’ best practices for server sizing and configuration as well as storage.

ESG Lab’s View

Increasing numbers of clients and applications make desktop management a daunting task for IT. The number of applications supported increases with organization size, compounding desktop management challenges for large organizations. With increasing numbers of corporate applications to support, ongoing maintenance and management tasks directly translate into considerable IT staffing requirements and costs. Like server virtualization, desktop virtualization is establishing a foothold in the data center among IT staffs looking to optimize their current PC environments.

HP’s P4000 SAN has a highly scalable, clustered architecture that simplifies management and allows customers to start at the level of capacity and performance they require today and grow their environments on demand. Additionally, it is easy to use and manage, while providing advanced features such as Network RAID, remote replication, and thin provisioning.

Customers can stretch their clusters to create multi-site SANs. We have seen storage systems that scale in this fashion with NAS and CAS products, but in ESG Lab’s opinion, HP P4000 is a leader in SAN attached true N-way clustered storage. ESG has long been a proponent of scalable clustered storage and believes it will become the dominant approach due to the compelling value it brings.

ESG Lab found that HP P4000 SAN performed well in a virtual desktop environment, providing easy provisioning and powerful integration of thin provisioning and Linked Clone technology to optimize capacity utilization. High availability functionality was also impressive, sustaining multiple failures while providing continuous access to attached virtual desktop users.

HP’s P4000 SAN systems delivered an easy-to-use, flexible, scalable, highly available, and highly efficient storage solution for VMware View customers. Matching in storage what VMware provides for desktops, HP P4000 supports thin provisioning and Linked Clones for creating large numbers of virtual desktops without the delay and cost of consuming actual storage for each clone. And because it is distributed by design, creating a disaster-resilient storage infrastructure is as easy as choosing which storage nodes to configure in each separate location.

Through hands-on testing, ESG Lab confirmed that the HP P4000 SAN provides a robust networked storage foundation for a virtual desktop architecture. With simple configuration, powerful desktop mobility, enterprise class availability, and near-linear scalability, the HP P4000 SAN enhances the intelligence and value of the VMware View virtual desktop infrastructure.

Appendix

[2] http://www.vmware.com/resources/compatibility/search.php?action=base&deviceCategory=vdm

[3] Configuration details are listed in the Appendix.

[4] Iometer configuration details can be found in the appendix.

This report documents ESG Lab hands-on testing of HP-branded Emulex Universal Converged Network Adapters (UCNAs) and explores how a converged infrastructure can enable users to realign traditional technology silos into adaptive pools that can be shared, optimized, and managed as a service.

Background

Organizations are in the process of transforming their data centers into environments that are able to better handle rapidly changing business requirements and unanticipated business needs. In order to meet these objectives, organizations are eliminating their monolithic legacy data centers and infrastructures and replacing them with consolidated, highly virtualized, flexible environments. Fibre Channel over Ethernet (FCoE), an industry standard protocol (ANSI T11) that maps Fibre Channel (FC) storage traffic over an enhanced 10 Gigabit (10Gb) Ethernet network, is a key technology that will enable that transformation.

FCoE technology has experienced an accelerated development cycle, going from concept to alpha, to beta, and now to second generation production-ready technology in just a few short years. Even more important, it has been tested, “certified,” and is supported by most major vendors. HP’s support will be a critical factor in FCoE adoption as organizations can now take advantage of the technology, knowing their trusted vendor will support it. ESG research[1] indicates that 2010 and 2011 should be significant years for FCoE adoption, as shown in Figure 1. IT professionals surveyed by ESG indicated that 10GbE is poised for 24% growth while FCoE usage looks to expand by 17%. Meanwhile, utilization of traditional Fibre Channel is flattening out.

The increased utilization and enhanced reliability of centrally managed FC storage networks have fueled a massive wave of storage consolidation in recent years. Driven by a set of challenges associated with the cost and complexity of servers, a growing number of organizations are using server virtualization technology to consolidate servers in the data center. As these trends continue, a new wave of I/O or network consolidation, supported by traditional Ethernet-based network traffic and FC-based storage traffic, has begun to take form.

This is to be expected, as the intent of FCoE is not to create yet another separate data center protocol but to merge FC with 10GbE and gracefully migrate from existing FC to Ethernet over time. With organizations looking to reduce operating expenses and accelerate business processes, a converged network enables rapid provisioning of new services and quick, seamless addition of throughput.

Fibre Channel over Ethernet

FCoE maps FC storage traffic over an Ethernet network with the goal of converging storage and networking traffic onto a single platform leveraging familiar management tools, security models, and processes. Switches and server-based adapters supporting FCoE are being evaluated by a growing number of organizations interested in the potential benefits of FCoE technology.

The Data Center Bridging (DCB) standard (a.k.a., Data Center Ethernet or Converged Enhanced Ethernet) was defined to enable the converged approach shown in Figure 2.  One of the key features of Enhanced Ethernet is its ability to differentiate between and prioritize different types of traffic sharing a common physical layer (a.k.a., Quality of Service, or QoS). The DCB standard also supports link-level flow control and end-to-end congestion management to meet the lossless performance requirements of mission critical applications currently relying on FC for networked storage connectivity.

The FCoE standard picks up everything from the FC standard—except the cabling and the physical interface—and places it on top of an DCB Ethernet network.  It was approved within the T11 committee in June, 2009.  FCoE has the potential to reduce data center complexity by reducing the number of cards, cabling, and network devices in the data center. Combining FC and Ethernet onto a single card, a Converged Network Adapter (CNA) replaces the need for separate network interface cards (NICs) and FC host bus adapters (HBAs) while allowing organizations to replace separate edge or top-of-rack Ethernet and Fibre-Channel switches with DCB multiprotocol switches supporting Ethernet and Fibre Channel connectivity.

Consolidation and virtualization are the key data center initiatives for transitioning from existing, segregated, LAN/SAN infrastructure to more efficient, converged architecture. The ultimate goal of network convergence is to provide the foundation for a truly agile end-to-end infrastructure while achieving significant reductions in capital equipment and operational expenses.

HP CN1000E Dual Port CNA

The HP CN1000E Dual Port CNA, seen in Figure 3, provides LAN and SAN connectivity over 10GbE network using FCoE and DCB functionality. The CN1000E has a PCIe Gen 2 X8 connection for server-side communication and two ports of 10Gbps Ethernet connectivity for network and storage connectivity.

HP CN1000E Dual Port CNAs and Emulex FC HBAs share a common FC driver stack and both can be managed with the Emulex OneCommand Manager application, this is especially important to administrators with an installed base of Fibre Channel HBAs. Both also support N-Port ID Virtualization (NPIV), which provides virtualized addressing and zoning for enhanced security in server virtualization environments. Leveraging a proven heritage and a common tool set, HP CNAs are designed to enable consolidation, energy conservation, and savings.

ESG Lab Validation

ESG Lab performed hands-on testing of the HP CN1000E at Hewlett Packard’s facility in Marlborough, MA. ESG tested the CNA’s ability to provide transparent network and storage connectivity for industry standard HP ProLiant servers. ESG Lab also looked at the management capabilities of the OneCommand Manager application.

Getting Started

The ESG Lab test bed consisted of servers, network switches, and storage as might be found in a typical data center.[2] Two physical servers were running Microsoft Windows Server 2008. One server was configured to represent existing FC attached servers in a traditional data center setup as seen in Figure 4, with an 8Gb FC HBA attached to an HP StorageWorks 8/24 FC SAN switch as well as a 1-Gb Ethernet NIC attached to an HP ProCurve 1-Gb Ethernet switch. The second server was configured with a single HP CN1000E Dual Port CNA attached to an HP StorageWorks 2408 Converged Network Switch for converged network and storage connectivity. Both servers were presented storage from an HP StorageWorks EVA 4400 FC storage array.

ESG Lab Testing

ESG Lab first examined the procedures required to incorporate the CNA and FCoE into an existing FC environment, including the configuration of both the CN1000E and the Converged Network Switch. ESG Lab referred to the HP StorageWorks SAN Design Reference Guide[3] periodically throughout the installation process. It contains detailed information about HP SAN architecture, including Fibre Channel, iSCSI, FCoE, SAN extension, and hardware interoperability.

First, ESG Lab downloaded the CNA driver package and then the OneCommand Manager application, installing them both on the DL385 G6 server where the CN1000E CNA was physically installed. The drivers and software are easy to find in the storage networking section of the HP website. The entire process, from the first mouse click on HP’s website through downloading and installing both the drivers and the OneCommand Manager—including rebooting after installing the drivers and software—took nine minutes. Once the driver package and OneCommand Manager were installed, OneCommand Manager was used to configure the CN1000E. As can be seen in Figure 5, the CNA was displayed along with the existing traditional FC HBA and showed both an Ethernet NIC and an FCoE adapter on port 1.

Additional tabs in the CN1000E provide configuration and verification of FCoE-specific parameters. The CEE tab, for example, shows Data Center Bridging state and mode as well as priority group properties as seen in Figure 6. Priority groups are configured on the Converged Network Switch and allow administrators to set bandwidth allocation for Ethernet and Fibre Channel traffic to ensure that both protocols can get the bandwidth required by applications.

Configuration of both FC storage and Ethernet networking was exactly the same as with traditional FC adapters and Ethernet NICs.  FC zoning of the HP 2408 DCB switch for FCoE connectivity was also familiar, using exactly the same process as for traditional FC SANs, as seen in Figure 7.

ESG Lab measured less than 15 minutes from the first mouse click to formatting FC SAN attached storage volumes on the FCoE attached server.

Ease of Management

The FCoE specification was defined with ease of integration with existing FC networks in mind. Bridged FCoE support can be used to add new servers to a converged 10GbE fabric while existing investments in FC switches and storage systems are preserved. Bandwidth allocation and management with priority groups is designed to ensure applications have the bandwidth they require, regardless of protocol.
HP takes ease of integration further with the HP SmartStart utility, delivered with HP ProLiant ML and DL  series servers and supporting HP ProLiant BL blade servers to provide step-by-step server deployment assistance. From array configuration and OS installation to the update of optimized ProLiant server support software, SmartStart helps administrators ensure a stable and reliable configuration.

ESG Lab Testing

To validate the interoperability and manageability of the CN1000E Dual Port CNA in an enterprise Fibre Channel SAN environment, ESG Lab ran several IO tests using the test bed shown in Figure 8.

ESG Lab tested dynamic bandwidth allocation using priority groups on the HP 2408 switch. Dynamic bandwidth allocation enables administrators to assign and re-assign bandwidth to applications on the fly, between the FC and Ethernet domains.  This provides more flexibility than is possible with separate FC and Ethernet networks, where additional bandwidth can only be achieved by adding adapters or HBAs and bandwidth cannot be shared across domains.

Priority groups were configured to guarantee 90% of available bandwidth to FC traffic and 10% to Ethernet. A SANBlaze RAM-based storage emulator was used to provide access to storage volumes at line-rate speeds over both 8-Gb Fibre Channel and 10Gb FCoE. An HP DL360 was used to provide 10Gb network connectivity, SAN traffic was generated using Medusa Labs test tools, and IP traffic was generated using the Iometer load generation utility.

First, ESG Lab ran a baseline test on the server with the traditional 8Gb FC HBA and a 1-Gb NIC installed. As seen in Figure 9, the speedometer from the Medusa Labs suite in the lower right showed that the FC HBA was reading at 773.27 MB/sec and the speedometer from the Iometer tool  in the Upper left showed that the 1-Gb NIC was reading at 112.82 MB/sec., both very near-line speeds.

Next, the same workloads were run on the CNA-attached server. The results are shown in Figure 10. FC traffic is getting almost precisely 90% of the total bandwidth and Ethernet traffic is getting the remaining 10%.

With the workloads still running, ESG Lab changed the priority group settings in the HP 2408 switch to re-allocate bandwidth. With two commands, allocation was changed to 50% for FCoE and 50% for LAN. Immediately after ‘Enter’ was pressed, the traffic pattern changed to what is seen in Figure 11, with no disruption or interruption to IO. At that point in time, FCoE traffic was getting 51% of the total bandwidth and LAN traffic was getting the remaining 49%.

Using a simple, single command, ESG Lab was able to dynamically re-allocate bandwidth from FCoE to LAN, with no disruption to service.

The Bottom Line

Organizations of all sizes are struggling to deliver IT services to meet the rapidly changing needs of the business. Data growth is unrelenting and IT budgets are constantly under scrutiny, so in order to meet increasingly demanding service levels, IT managers are looking for opportunities to consolidate. In order for this to be effective, IT not only needs to reduce capital costs, but more importantly needs to reduce operational costs, enhance scalability and performance, and improve cycle times. Server virtualization is one technology that meets these requirements and it therefore continues to be a top IT initiative. HP’s converged network offering also delivers these benefits by leveraging consolidated management using simple, familiar tools, reduced  hardware and cabling requirements, strong performance, and lower power and cooling requirements.

ESG Lab created a simple cost of implementation model based on a typical data center environment with a requirement to add 100 new servers to an existing environment with a 4Gb FC SAN infrastructure and 1Gb LAN distribution.  ESG Lab chose to make the comparison using 4Gb FC because it fits better with where most users environments are today.  In the data center, 4Gb FC is much more common than 8Gb.

Costs were broken down into two major categories:  capital expenditures including switches, NICs, and HBAs as well as cables, and operational expenses including power and cooling as well as cable installation.  Hardware costs were calculated using online pricing from HP and other major vendors, power requirements were taken from hardware manufacturers’ specification sheets, and operational expenses were estimated based on interviews and conversations with IT managers responsible for managing large enterprise data centers.[5]

Figure 12 shows the relative capital expense a typical end-user would incur to implement 100 servers with a traditional combination of 4Gbps Fibre Channel SAN and Gigabit Ethernet infrastructure versus implementing 100 servers using a single, converged network with 10GbE and FCoE.

Figure 13 shows the annual power and cooling expenses for the switching and network adapters in both scenarios.

Table 1 shows the detailed results of ESG Lab’s calculations, with cost savings broken out by category.

What the Numbers Mean

• Consolidating to a 10Gb converged network infrastructure resulted in significant cost savings in both capital and operational expenditures compared to a traditional LAN/SAN infrastructure.
• Operational expenses such as power/cooling and cable plant maintenance showed significantly higher cost savings over the cost of switches and adapters due to the large reduction in cabling and number of switches. As the environment scales over time, these savings can really add up.

ESG Lab Validation Highlights

• The HP CN1000E Dual Port CNA was recognized by both OneCommand Manager and Windows as a FC HBA and a 10GbE NIC, without any special software or configuration required.
• The CNA used the same lpfc drivers as all standard Emulex FC HBAs.
• The management experience administering the CNA is essentially identical to any HP-branded Emulex FC HBA.
• The CNA was able to participate in a Fibre Channel fabric with 8Gb FC HBAs and share the same storage array without any compatibility or interoperability issues.
• The priority groups feature on the HP 2408 Converged Network Switch was able to allocate bandwidth between FC and Ethernet applications precisely and on the fly, with no disruption or interruption in service.
• ESG Lab estimated that an organization can use FCoE to add new servers to their existing FC SANs with minimal added cost of acquisition, with no additional administration costs, and with fewer cables, adapters, and switches to power, cool, and manage.

Issues to Consider

• While the T11 committee has approved the FCoE standard and sent it up for public review, the DCB or Enhanced Ethernet standards are still being ratified. Having said that, forward thinking IT managers with existing investment in FC fabrics would be wise to begin investigating this technology sooner rather than later.
• Upgrading from traditional FC to FCoE in one fell swoop is potentially problematic, requiring replacement of SAN and network switches as well as HBAs. The good news is that FCoE, deployed in its most likely scenario—when adding new servers—integrates seamlessly into existing SAN and LAN environments without having to replace existing FC infrastructure. As that SAN and LAN infrastructure ages and needs to be replaced, users migrate to FCoE organically.
• While administration and troubleshooting should be easier with a single console for both FC and Ethernet connectivity as well as fewer cards and cables, there are potential conflicts of IT culture: joining the FC and Ethernet networking teams together.

The Bigger Truth

As organizations continue to transform their data centers and server virtualization continues to proliferate, servers, storage, and networking devices must be added to keep pace with business demands. As a result, businesses are spending upwards of 70% of their IT budgets on operations.

In a recent ESG survey,[6] 54% of IT managers polled indicated their organizations would be basing new technology decisions based on their ability to reduce operational costs and 42% indicated improving business processes would be their most important criteria. Converged networks should be an attractive solution for IT departments looking to reduce footprint and complexity in the data center while streamlining administration. It should be noted that in this same survey, reducing capital costs went from tied for second place in 2009 down to fifth place in 2010. ESG has had numerous conversations with organizations that validate this research, with some organizations stating they can get all the capital they need—provided they can show it will help to reduce operational expenses.

ESG Lab found that a converged infrastructure enables users to reduce both capital and operational expenses while increasing flexibility and service levels with a simplified architecture. Fewer network switches, cards, and cables, with more efficient silicon technology, result in less power being consumed and less heat generated while administrative burdens are reduced with far fewer devices and connections to manage.

Leveraging familiar tools and processes, DCB and FCoE provide a smooth upgrade path for the migration of storage traffic from an isolated FC storage network to a converged Ethernet fabric. It is ESG Lab’s opinion that FCoE is a compelling technology and potentially a better fit than iSCSI for organizations with significant existing investment in Fibre Channel. It enables companies to retain existing FC infrastructure, keeps existing FC management tools in place, provides the same level of performance guarantees, and has the potential of reducing costs. This final consideration of cost is a significant variable in the FCoE adoption equation. Given the present economy and the pressure IT is under to reduce costs, compelling FCoE pricing could be used to accelerate adoption faster than any marketing pitch or certification.

ESG Lab has seen FCoE in action, with HP CN1000E CNAs providing converged server connectivity to a LAN and a SAN. Running alongside HP-branded Emulex HBAs and managed from the familiar Emulex OneCommand Manager application, HP CNAs worked with field-proven drivers and tools. Tens of thousands of customers have taken advantage of HPs decades of IT experience to deploy rock solid FC and iSCSI SANs. HP’s CNA provides the flexibility to choose not only the right protocol, but also the right performance and price point to address changing business needs.

ESG believes that organizations should take a good, long look at this architecture and consider how they could benefit from building highly flexible compute resource pools that can adapt to multiple network technologies—without requiring significant moves, adds, or changes. Simply put, network convergence makes sense. If you don’t have to have more than one network, why would you? This type of cost savings and flexibility should be well received by organizations planning their data center transformations.

Appendix

[2] Complete configuration details can be found in the Appendix.

[3] http://h20000.www2.hp.com/bc/docs/support/SupportManual/c00403562/c00403562.pdf?jumpid=reg_R1002_USEN

[4] Source: ESG Research Report, Global Green IT Priorities: Beyond Data Center Power and Cooling, November 2008.

[5] Assumptions and Parameters can be found in the Appendix.

[6] Source: ESG Research Report, 2010 IT Spending Intentions Survey, January 2010.

[7] As of Dec. 09 according to the US Energy Information administration http://www.eia.doe.gov/cneaf/electricity/epm/table5_6_a.html

The Challenges

Faced with exploding data growth and increasing demands thanks to server consolidation and virtualization as well as a need to better protect and manage shared information assets, IT managers in organizations of all sizes are increasingly leveraging a shared storage approach using SAN technology. In the enterprise, this has meant Fibre Channel switches and directors, an option often out of reach of smaller businesses. In fact, when ESG asked IT professionals in small- to medium-size businesses to  name the business initiatives which would have the greatest impact on IT spending decisions over the next 12-18 months, the number one response was cost reduction, with simplification of business processes and risk management a close second and third.[1]

The Solution: Cisco MDS 9148 Multilayer Fabric Switch

The MDS 9148 is a 48-port, non-blocking Fibre Channel switch that combines the NX-OS operating system used in Cisco’s Nexus series of data center Ethernet switches and MDS 9000 SAN switches and directors with configuration wizards for simplified deployment and management. It is offered with a pay-as-you-grow licensing model supporting from 16 to 48 ports of 8-Gbps storage area network (SAN) connectivity.

Priced as an entry level switch, the MDS 9148 provides many of the enterprise features that are included in the MDS 9500 Series Multilayer Directors and MDS 9200 Series Multilayer Fabric Switches:

• 8-Gbps, non-blocking throughput on all 48 ports.
• FlexAttach technology to enable transparent server deployment without the need to reconfigure the SAN.
• N-Port Virtualization (NPV) technology and N-port ID virtualization (NPIV) support for virtual environments.
• VSAN support for security and fault isolation.
• PortChannels for aggregating up to 16 physical ports into one logical channel
• In Service Software Upgrade to enable firmware upgrades without removing the switch from service.
• Dual hot-swappable power supplies and fans to support high availability.
• RADIUS and TACACS+ port security, Fibre Channel Security protocol, Secure FTP, Secure Shell Version 2, and role-based access control for comprehensive security controls.
• The 9148, like all MDS switches and directors, blocks corrupted and malformed frames from entering a fabric –protecting servers and applications.

ESG Lab validated the performance and enterprise-class capabilities of the MDS 9148 during a day of hands-on testing at Cisco’s campus in San Jose, California.

Performance/Scalability

Like  the previous generation MDS 9124, the Cisco MDS 9148 is a fully non-blocking fabric switch. ESG Lab tested performance to  validate that the 9148 was able to pass traffic at 8 Gbps on all 48 ports simultaneously. Testing was executed using three 16-port JDS Uniphase load testers connected to all 48 ports of one MDS 9148 at 8 Gbps as seen in Figure 2.

The load tests were executed using JDSU Xgig Maestro software, which was used to verify the rate at which all channels were transmitting and receiving data as well as latency. The load testers were configured to transmit on odd-numbered ports and receive on even-numbered ports at 99.9% of wire speed.

ESG Lab Testing

Three tests were run to validate performance in different configurations. The first test was run with full size frames to test throughput and was performed with a fixed payload size of 2,112 bytes. ESG Lab examined every port during the test and confirmed that all odd-numbered ports were sending and all even-numbered ports were receiving at just over 800 MB/sec. or 7.81 Gbps on each port simultaneously. It is important to note that this represents 99.8% of wire speed.  The actual throughput driven by the load tester is rounded down from the configured value of 99.9% due to internal calculations.[2] Stated another way, ESG Lab confirmed that a single MDS 9148 delivers an aggregate throughput of more than 131 Terabytes per hour across all ports.  During this first phase of performance testing, the MDS 9148 delivered perfectly consistent low latency of 3.0 microseconds per frame as shown in Figure 3.

The throughput test was repeated with a minimal, fixed payload size of 24 bytes. Small frame tests show how efficiently the switch can process the Fibre Channel protocol. Even though the payload size was almost 100 times smaller than the first test, all ports were sending or receiving at an impressive 552 MB/sec. with a consistent 1.0 microsecond average latency per frame. In this test, each port was processing 9.66 million frames per second. That’s an impressive 463.68 million frames per second across the whole switch.

The next two tests were designed to evaluate fair traffic allocation in the switch. A fan-out test was conducted with 15 ports transmitting to one receiving port. This test was performed to confirm that the switch would effectively divide the load equally between the fifteen simulated server ports. Each of the 15 transmitting ports was attempting to send 800 MB/sec. of data to the target port. The MDS 9148 allowed an actual load of 53.4 MB/sec. on each of the 15 transmitting ports with a full 800 MB/sec. load on the target port.

The final test was a head-of-line blocking test. Head-of-line blocking is a condition where traffic waiting to be transmitted to a congested destination port prevents or blocks traffic destined elsewhere from being transmitted. In this test, the JDSU Xgig load tester was configured with output from port 1 sent to port 2, while port 3 was configured to send output to both ports 2 and 4. The result was an even distribution across ports 2 and 4 of the traffic coming from port 3. The traffic from port 1 was throttled to 50% to accommodate the load from port 3.

Flexibility and Ease of Management

Cisco has long offered the ability to configure multiple levels of administrative users with role-based access. FlexAttach is a feature that takes advantage of this capability to enable server administrators to move, add, or change server hardware without reconfiguring the SAN. FlexAttach uses zones to associate Cisco Virtual Port World Wide Names (VPWWN) with storage volumes. When a server is attached, its Host Bus Adapters (HBAs) are associated with the appropriate Cisco VP WWN. This functionality allows SAN administrators to protect the configuration of the switch fabric, while still allowing server administrators to replace hardware without assistance from the SAN administrators.

ESG Lab Testing

ESG Lab used the test bed illustrated in Figure 4 for the remainder of testing detailed in this report. One Windows 2003 server with a Qlogic QLA 2342 HBA installed was attached to a single MDS 9148 switch which was in turn attached to a second MDS 9148 via two inter-switch links (ISLs). The second MDS 9148 was attached to an Adaptec SANbloc-S50 disk enclosure.

ESG Lab utilized a two-step configuration wizard launched from Cisco’s Fabric Manager GUI to configure the ‘Edge’ MDS 9148 to enable FlexAttach on each port. After the configuration was completed, the new Cisco VPWWN assignment was seen under the server’s HBA. Figure 5 illustrates the new configuration, showing the server’s physical Qlogic HBA and the virtual Cisco HBA configured on the SAN.

A hardware failure was simulated by unplugging the attached server from the SAN fabric. A new server was then introduced to the MDS 9148 and the Fabric Manager program was launched using a login that had been previously defined as a server administrator. This role had limited rights to the SAN fabric when compared to the SAN administrator and was only allowed to add, move, or change specific servers attached to specific switches in the SAN. Using the FlexAttach functionality as shown in Figure 6, ESG Lab was able to associate the new server’s HBA WWN with the VPWWN owned by the replaced server. Using Windows Disk Administrator, the new server was verified as able to mount the correct disks without changing zoning or LUN masking.

Manageability, Serviceability, and Cost Efficiency

A number of powerful, enterprise-class capabilities enhance the manageability, serviceability, and cost efficiency of the Cisco MDS 9148 Multilayer Fabric Switch:

• Manageability: N-Port Virtualization, Configuration Wizards
• Serviceability: Online firmware upgrade, PortChannels, redundant hot swappable fans and power supplies
• Cost-Efficiency: Port upgrades from 16 to 48 ports in eight port increments, no additional licensing fees for enterprise-class features

This section documents ESG Lab testing of three of the enterprise-class capabilities of the MDS 9148: N-Port Virtualization, PortChannels, and In Service Software upgrades.

N-Port Virtualization

By default, Cisco Nexus 5000 and MDS Series switches operate in fabric mode. In fabric mode, each switch that joins a SAN is assigned a unique domain ID. Each SAN (or VSAN) supports a maximum of 239 domain IDs, so each SAN has a limit of 239 switches. In large scale, edge to core topologies, SANs may need to grow beyond this limit. N-port Virtualization (NPV) significantly reduces the number of domain IDs needed in a fabric by eliminating the need for a domain ID on edge and top of rack switches.

In NPV mode, the edge switch relays all traffic from server-side ports to the core switch. The core switch provides F-port functionality (such as login and port security) and all Fibre Channel switching capabilities. The edge switch appears as a Fibre Channel host to the core switches and as a regular Fibre Channel switch to its connected devices. As a result, the devices attached to the NPV-mode switches get their domain IDs from the core switch.

ESG Lab configured NPV using a wizard launched from the Cisco Fabric Manager application. This process automatically completed the conversion of two MDS 9148 switches to NPV devices. After the wizard was completed, both switches shared the same domain ID, as shown in Figure 7.

PortChannels

PortChannels are used to aggregate up to 16 physical inter-switch links into a single logical path. This provides optimized bandwidth utilization across all links in the PortChannel. When the other end of the PortChannel is a modular chassis, such as an MDS 9500 director, the bundle can include any port from any ASIC in the chassis, ensuring that the channel remains active even in the event of a module failure.

ESG Lab created a PortChannel between edge and core MDS 9148 switches using two inter-switch links on ports 13 and 25 on both switches.

The industry standard IOMETER workload generator was used to create read and write traffic transmitted from two servers attached to the edge MDS 9148 switch across the SAN fabric to two disk volumes on an array attached to the MDS 9148 core switch. Port 13 on the core 9148 switch was then disabled to simulate a port failure. The load was taken on by the remaining active link (port 25) and no disruption in service was experienced on the server.

As illustrated in Figure 8, the red and yellow bars show server traffic while the blue and gray bars represent traffic on ports in the PortChannel. Note that when Port 13 (represented by the blue bar) was disconnected, traffic was rerouted through Port 25 (represented by the gray bar) and traffic to and from the servers (the red and yellow bars) was unaffected.

In Service Software Upgrades

During the service life of any switch, it is not unreasonable to assume that firmware will, at some point, need to be updated to either fix a bug or enable an enhanced feature. Cisco’s In Service Software Upgrade (ISSU) provides the ability to perform a full NX-OS image upgrade without having to take the switch out of service.

To test both the functionality and the non-disruptive nature of ISSU, the industry standard IOmeter workload generator was used to perform read and write IO from a SAN-attached server while the Software Install Wizard was launched from Cisco Device Manager. The entire software installation process took less than four minutes to complete. IOmeter continued running throughout the entire process with no errors and no loss of connectivity to storage.

Figure 9 shows the Cisco MDS traffic monitor while the software install wizard was running. The red bar shows read and write traffic being generated by the server on port 3, while the yellow and blue bars represent traffic to and from the storage devices on ports 13 and 25. ESG monitored this display, as well as the IOmeter console, while the software upgrade was running to verify that there was no disruption to service.

The Bigger Truth

This ESG Lab review has confirmed that the NX-OS based Cisco MDS 9148 fabric switch is a very effective solution for small- to medium-size businesses looking to migrate from DAS to high availability SANs without sacrificing the enterprise-class features needed in large scale edge-to-core deployments. Using a switch-on-a-chip architecture, which ESG Lab has proven can scale up to 48 ports of non-blocking, 8-Gbps bandwidth, the MDS 9148 provides a high performance, 16-port entry for a SAN that can be easily upgraded to support future growth. First time SAN users will appreciate the user-friendly configuration wizard which guides an inexperienced user through initial set up in minimal time.

Cisco is making good on its strong commitment to support the small- to medium-size business (SMB) market with full featured products like the MDS 9148 just as it did with the MDS 9124. Enterprise-class features—including VSANs, security, non-disruptive firmware updates, and NPV—provide customers in this market segment with the versatility and capability they need to grow with their businesses. More importantly, all these features will continue to be delivered as part of the standard package, not as incremental costs.

Enterprise customers should also consider the MDS 9148 as a good choice for top-of-rack architectures enabling a core-to-edge solution, especially with existing MDS Directors as they will share the same NX-OS. Its wizard-based configuration options will make adding switches and servers to remote sites or the edge a simple process that will not require experienced SAN professionals’ time. The price points could make it more attractive to include lower tiered stranded servers into the SAN environment.

ESG would recommend that Cisco continues to work with its partners to develop simple, preconfigured “SAN in a box” solutions to further aid adoption. These end-to-end solutions help reduce the complexity and ease installation. Ideally, leveraging wizard-driven configuration for the entire solution would simplify the solution implementation process even further.

Pricing is expected to be competitive for entry level configurations and ESG expects Cisco to aggressively position this product to drive market share in this space: Cisco only competes in a space in which it can be a market leader. Including enterprise-class NX-OS features such as Fabric Manager, NPV, VSANs, security, PortChannels, and comprehensive diagnostics in the base price of the switch will certainly be a differentiator. Priced correctly, ESG expects the MDS 9148 will gain significant market traction given the appropriate sales channels.

ESG Lab found that the MDS 9148 is a rock-solid, feature-rich, 8-Gbps multilayer fabric switch that cost effectively delivers the benefits of SAN-attached connectivity to small- and medium-size businesses and enterprises looking for   affordably powerful edge connectivity.

[2] “MB/sec.” is calculated as 1024*1024 or 1,048,576 bytes per second. Gbps is calculated as 1024³ or 1,073,741,824 bits per second.