Nirvanix had some pretty impressive wins in 2011, and has done a good job getting its customers to go public.  It has been roughly a year since the new management took over, and their diversification strategy seems to be working.  Nirvanix sells both software and storage services.  Notable wins in 2011 include:

• Relativity Media–next gen studio behind hits like “Immortals” and “Limitless”; will leverage NVX public cloud for content collaboration to bring movies to completion faster.
• NBC Universal is deploying 2+ Petabytes (videos, photos, movies; leveraging Nirvanix public cloud for content archival and collaboration).
• Advocate Healthcare has 500TB (healthcare content files; NVX public cloud).
• USC deployed 8.5 Petabytes (Videos, Photos; USC will also resell NVX Private Cloud as the USC Digital Repository).
• Cerner Healthcare for 2+ Petabytes (for PACs, radiology, clinical systems, patient records; Cerner will also resell NVX Private Cloud as Cerner Skybox Public Cloud).
• IBM is deploying a Multi-Petabyte Global Private Cloud; NVX is essentially building a replica of its Cloud Storage Network for IBM to sell as its own public cloud–IBM SmartCloud Enterprise.
• DRFortress is offering an onsite physical node federated with Nirvanix public cloud, providing its customers with a hybrid cloud storage deployment.

The first three are true public cloud users that are leveraging Nirvanix to accelerate business (because, as Steve D. mentioned in his blog last month, doing all the storage stuff yourself stinks, or something to that effect).  But the rest are resellers as well as users.  Now, this is a pretty impressive list – names we know and love.

I’ve actually talked to a few of the resellers and there seems to be strong momentum in the channel.  I don’t think it would be too risky to say that the strategy Nirvanix management laid out a year ago and has been executing against since is paying off for them.  Granted, Nirvanix adoption is only a microcosm in the larger cloud universe – but it is proving that the business model can work and there is money in IaaS.  Rumors of Nirvanix’s demise were swirling a mere 18 months ago, rumors which the new team seems to have silenced.

Looking forward to 2012, it would not be surprising to see Nirvanix gobbled up by a bigger player who can leverage the current cloud delivery network infrastructure, cloud software, and hybrid cloud model to offer both public and virtual private cloud services.  There are a lot of variables – a lot depends on the economy, and it is sure to be a shaky year because of the elections.  But Nirvanix is one I will be keeping an eye on.

You can read Terri’s other blog entries at IT Depends.

Storage Center from Dell Compellent was designed to provide efficient, agile, and resilient networked storage with a goal of reducing storage cost and complexity while meeting the growing needs of the business.

This report documents hands-on testing of Storage Center version 5.4 with a focus on the capacity, recoverability, and performance efficiency of the Compellent Dynamic Block Architecture. Testing covered Compellent’s approach to multi tiered storage, referred to as Data Progression, as well as its deep integration and manageability in VMware virtual server environments and automated migration of virtual machines between sites with the recently-added Live Volume feature. One of the key goals of this report is to highlight the capabilities of Compellent Data Progression as compared to emerging sub-LUN tiering solutions from other leading storage vendors.

Click here to read the full report.

Introduction

Storage Center from Dell Compellent was designed to provide efficient, agile, and resilient networked storage with a goal of reducing storage cost and complexity while meeting the growing needs of the business. This report documents hands-on testing of Storage Center version 5.4 with a focus on the capacity, recoverability, and performance efficiency of the Compellent Dynamic Block Architecture. Testing covered Compellent’s approach to multi‑tiered storage, referred to as Data Progression, as well as its deep integration and manageability in VMware virtual server environments and automated migration of virtual machines between sites with the recently-added Live Volume feature. One of the key goals of this report is to highlight the capabilities of Compellent Data Progression as compared to emerging sub-LUN tiering solutions from other leading storage vendors.

Background

IT departments continue to struggle with the costs of storage, which includes not only the cost of acquisition but also the ongoing operational expenses associated with storage management. Growing demand for increased capacity and the need to provide assured availability and flexible recovery scenarios are challenges as well. As shown in Figure 1, a recent ESG survey indicates that rising costs, rapid data growth, and data protection are at the top of the list of challenges associated with meeting the storage needs of applications.[1]

Keeping up with the growing storage needs of virtual server environments and the cost and complexity of data migrations are critical concerns as well. And last, but not least, IT managers are looking for products and tools which can automate the management, optimization, and capacity balancing of storage resources associated with ever-growing information assets.

A storage solution with an architecture that dynamically automates data placement and protection at the block level is needed to address each of these challenges as it stores and protects application data using the right tier of storage, at the right time, at the right price.

Storage Center 5.4

Storage Center is a SAN attached storage solution that uses a combination of intelligent software, a pair of clustered storage controllers built from industry-standard server hardware, host  interface controllers (e.g., Fibre Channel, iSCSI), and one or more drive enclosures to store and protect an organization’s data. Drive enclosures can be populated with multiple tiers of storage to meet a variety of price and performance requirements (e.g., SAS and SATA), with the ability to “mix and match” storage technologies and interfaces in a modular fashion.

Along with a fourth generation controller boasting a 25% IO performance increase and more controller memory and PCI-e slots, Storage Center 5.4 also offers support for 6 Gbps SAS drives, 2.5” SAS enclosures for higher density, and support for the latest host connectivity options (10 GigE iSCSI and FCoE).

Compellent, which was recently acquired by Dell, was founded in 2002 by a visionary team of storage industry veterans. Storage Center was architected from the ground up with a powerful page-based architecture that automates the movement and protection of data at the block level data using capacity residing on different tiers of storage. As shown in Figure 2, policies are used to automatically use higher performing “tier-1” storage for write data and frequently accessed read data. Dell’s Data Progression functionality automatically migrates infrequently accessed blocks of data to a more cost effective tier of near-line storage. Replays (a.k.a., snapshots) are automatically placed on lower tier(s) of storage for cost effective recovery.

Since Compellent first began shipping systems with its innovative Data Progression feature in 2003, a number of leading storage vendors have begun to adopt a similar approach with a technology that has since become known as sub-LUN tiering. At that time, the Dynamic Block Architecture was also used to reduce the cost of disk capacity via a thin provisioning feature which Dell refers to as Dynamic Capacity. Once again, Compellent was one of the first to introduce what may be better known by many as thin provisioning. The Dynamic Block Architecture also enables a number of powerful capabilities including WAN efficient replication of recovery data between sites (Remote Replays) and Live Volume which create a single view of application data residing on a pair of Compellent Storage Center solutions deployed within the same—or different—data centers.

ESG Lab Validation

ESG Lab performed hands-on evaluation and testing of Storage Center version 5.4 at a Dell facility in Eden Prairie, Minnesota. Testing was intended to validate the Dynamic Block Architecture’s delivery of efficient, agile, and resilient storage services. ESG Lab looked at several aspects of the Storage Center solution, focusing on:

• Capacity efficiency with Dynamic Capacity
• Performance acceleration with Data Progression
• Efficient disk-based recoverability with local and remote snapshots (Remote Instant Replay)
• Storage and server virtualization with VMware and Microsoft Hyper-V integration
• Advanced agility and recoverability with Live Volume

Getting Started

The test bed used during the ESG Lab Validation is shown in Figure 3. A Microsoft Windows physical server and a VMware vSphere virtual server were connected via 8 Gbps Fibre Channel to a Storage Center solution with a mix of SSD, FC, and SAS drives. The test bed specifics can be found in Table 1 of this report’s appendix, with nearly 80% being large but lower-performing “tier 3” drives and the rest being tier-2 or tier-1 to quantify the performance benefit after tiering up to faster storage with Data Progression and the efficiency of snapping down to lower cost storage during Replay testing. A similar Storage Center configuration located in a simulated remote site was used during remote Replay and Live Volume testing.

Capacity

Compellent’s page-based Dynamic Architecture is used to implement the Dynamic Capacity feature generally known in the industry as thin provisioning. Dynamic Capacity is used to quickly and efficiently allocate storage on demand without consuming blocks that would otherwise be wasted before a server actually used them. As opposed to legacy storage systems that have recently been upgraded to support thin provisioning, Dynamic Capacity was built into the Compellent architecture from the beginning and is the default behavior for storage provisioning.

Figure 4 illustrates the difference between Compellent Dynamic Capacity and traditional provisioning. In this example, a new 250 GB volume has been created and presented to a server. In both cases, the server sees 250 GB of available capacity. With traditional provisioning, that capacity is exclusively dedicated to the server, regardless of when or how much the server actually consumes. On the right, Dynamic Capacity presents the same capacity to the server, but consumes only the blocks that have been written by applications. In this example, a small fraction of the 250 GB in usable capacity has been used. ESG research and conversations with a number of early adopters of thin provisioning indicates that customers have reduced the cost of storage capacity by a factor of 50% or more compared to traditional provisioning method.

ESG Lab Testing

ESG Lab tested Dynamic Capacity using a simple deployment of file server storage. A 249 GB volume was created using the web-based Storage Center management console and attached to a server running the Windows Server 2008 R2 operating system. A file system was created and a drive letter was assigned with the Microsoft Disk Administrator utility. A Windows Explorer properties view of the drive is shown in Figure 5.

Figure 6 shows that the actual disk space consumed for the 249 GB volume is only 138 MB.

Performance

Data Progression leverages the Dynamic Block Architecture to move individual blocks of data to different tiers of storage based on predefined or custom policies. Data Progression is managed at the page level, with each page being composed of one or more blocks of data (the default page size is 2 MB). These pages can then be managed for performance and other goals by intelligently placing them on the most appropriate type of storage (e.g., SSD, SAS, SATA), RAID-levels (e.g., RAID-10, RAID-5), and even track-placement on a given spindle (Fast Track).

The Compellent Dynamic Block Architecture stores write data on a pool of blocks residing on the fastest performing tier of storage. As illustrated in the example shown in Figure 7, writes are stored on a pool of storage protected with a high performance RAID-10 algorithm. Reads are serviced from that same group of drives until the Data Progression algorithm kicks in (the default is once per day at 7 PM).

Data Progression migrates infrequently used blocks to a more cost effective tier of storage based on policies managed by an administrator. In this example, data progression migrates first to RAID-5 storage in the tier-1 pool (note that this novel approach provides a cost effective way to avoid the performance penalty associated with RAID-5 writes). Later, the automated Data Progression algorithm moves infrequently accessed data to a cost effective second tier of storage. If and when data becomes “hot” again, Data Progression reverses the process and migrates the blocks back to the higher performing tier.

ESG Lab Testing

ESG Lab application workloads and the industry standard open source IOmeter utility were used to quantify the performance benefit that can be achieved with Data Progression.[2] An online transaction processing (OLTP) database workload that simulates the activity of a multi-user Microsoft SQL Server application was tested. A virtual desktop infrastructure (VDI) workload that simulates a VMware View virtual desktop infrastructure using the linked clone method to share operating system “gold” images was also tested.

Data Progression testing was performed with an 8 Gbps Fibre Channel attached Compellent Storage solution that was configured with three tiers of storage:

• Tier 1: Four 200 GB SSD devices
• Tier 2: Forty-three 150 GB 15K RPM SAS drives
• Tier 3: Eleven 2 TB 7.2K RPM SATA drives

Application workloads were tested after Data Progression had run with a goal of measuring the performance boost that can be achieved with the fully automated page-level tiering algorithms built into Compellent Storage Center version 5.4. The results are summarized in Figure 8 and Figure 9.

ESG Lab observed what Dell referred to as common guidance in that only a relatively small amount of the higher‑performing storage was added to the environment. In conversation, ESG learned that most Compellent customers solve their capacity issues by purchasing only additional lower tier drives. Similarly, those same customers solve their performance issues by buying a smaller high-performance drives for a new or augmented top tier. This is a welcome contrast to the legacy approach of having to relocate entire data sets between storage arrays or tiers as workload performance needs change over time.

What the Numbers Mean

• The response times for the database application were 60% faster when moved from tier-3 SATA devices to tier-1 SSD devices.
• Faster response times at the IO level, which are shown as less in Figure 8, are magnified at the application level where each database transaction typically requires multiple IO requests.
• Moving the virtual desktop workload from tier-2 SAS drives to tier-1 SSD drives increased the number of IOs per second by 98%.
• ESG Lab measured IO activity of 20 IOs per second (IOPs) for an ESG analyst over the course of an eight hour work day. Using that value as representative of a typical knowledge worker, a recorded IO rate of 5,453 IOPS can support up to 272 heavy VDI users—98% more than was achieved with tier-2 SAS drives.

As notable as the numbers are, the experience was equally noteworthy. After configuring the IO tests with the initial storage, ESG Lab found it very easy to enable a higher-performing tier of storage. As the Compellent algorithms took effect, ESG was able to observe the increased performance as the faster storage began being utilized.

Recoverability

The Compellent Dynamic Block Architecture is used to create disk-based point-in-time Replays (a.k.a., snapshots). Administrators use Replays to perform non-disruptive backups of production data, as seen in Figure 10, as well as for application testing, development, and fast recovery of deleted files or from corrupted data. In this example, a file created at 11:55 AM was recovered from disk using a Replay Volume created at 12:10. From an end-user standpoint, the Replay was used to dial back in time and recover the file using a point-in-time image created before the corruption.

Compellent’s Dynamic Block Architecture not only enables blocks to seamlessly move between media tiers for optimized performance and cost, but is also the enabling technology for local and remote Replays. The Dynamic Block Architecture enables capabilities that are often not supported in snapshot implements within legacy storage solutions:

1. Replay data is stored on a lower cost tier of storage compared to the pool where the primary data resides. For example, a volume residing on high speed SAS drives can be protected with Replay capacity that resides on denser, more affordable SATA drives.
2. Replay capacity is delivered on demand from a pool of available blocks. Replay capacity does not have to be reserved up front.
3. Remote Replays send only the latest changes over the network to a remote Storage Center. This reduces WAN bandwidth requirements as it creates a cost effective remote mirror for disaster recovery.

ESG Lab Testing

ESG Lab used a Compellent Replay to recover from simulated data corruption. Specifically, ESG Lab created a volume from a replay so that the production server would continue as is, while a mountable version of the volume from a previous point in time was used for the recovery.

Because a Replay is simply a different list of pointers to blocks within the same Dynamic Block Architecture, a previous point in time consumes almost no space on its own, as seen in Figure 11 where both the production volume and the replay volume are visible in the left-hand tree, but the Replay Volume only consumes only 78 MB of actual disk space.

Figure 12 shows the same Replay volume as a separately mountable and usable resource on a Windows Server.

The key here is to recognize that the Replay volume is in fact “just pointers,” but Windows couldn’t tell the difference. While Figure 11 shows the replay volume made up of only 78 MB of overhead, Windows sees an entirely separate 250 GB volume. For this test, ESG Lab was able to easily create a Replay volume and then mount it to the original production server so that a previous version of a file could be accessed. The same capability can be used to create new server volumes from a master-image scenario.

Virtualization

According to recent ESG research,[3] 70% of organizations are using more than one server virtualization platform. Just as IT has managed and maintained a heterogeneous operating system environment, evidence also suggests that multiple hypervisors will be deployed. IT is heterogeneous by nature, so this makes complete sense. All hypervisors essentially do the same job, but licensing fees for proprietary products are driving enterprises to adopt a second choice. The increasing maturity of Citrix XenServer and Microsoft Windows Server 2008 servers running Hyper-V are grabbing enterprise attention, but VMware is not necessarily being displaced where it exists.

A natural pairing to Compellent’s virtualization of storage is to combine Storage Center with server virtualization platforms, including both VMware and Microsoft Hyper-V. This pairing is done from two perspectives:

• A hypervisor-aware perspective from within Storage Center
• Compellent-embedded information within server virtualization management tools

ESG Lab Testing

Figure 13 shows the multiple Storage Centers used during testing and the range of operating systems and virtual server hypervisors connected to them. In the expanded view on the right side, one can see ESXi and Windows Server 2008 R2 (Hyper-V) servers, each with multiple storage LUNs attached to each hypervisor.

ESG Lab tested Compellent’s integration with VMware ESX and vSphere 5.0 by monitoring and managing several typical storage functions from within the vCenter interface. It should be noted that similar integration was observed through the Microsoft PowerShell command language that provides scripting control for Hyper-V environments; however, the integration was best observed in the graphical vSphere UI.

ESG Lab observed many of the same provisioning activities and interfaces done earlier with physical servers but using the vSphere Client interfaces. Specifically, Figure 14 shows similar data to what was observed in Figure 6, related to a particular volume’s consumption of storage across multiple tiers within the Storage Center.

What is most notable about the integration is the “Compellent View” tab in the VMware vSphere management console as shown in Figure 14. In this example, the Compellent attributes of the storage associated with a virtualized server are shown. Several common monitoring and management tasks were executed from this VMware management console.

To further test the integration, ESG Lab attempted to manage the ESX storage from its normal view within the vSphere Configuration Tab, as seen in Figure 15.

Specifically, ESG Lab tested the ability to invoke a “Replay,” commonly referred to as a snapshot, from within vSphere. To accomplish this, ESG Lab selected an ESX cluster in the left-pane of vSphere. After selecting storage in the hardware selection pane, a right click on a volume revealed the Compellent sub-menu of commands shown in Figure 15.

ESG Lab anticipates that VMware administrators will be highly impressed with and empowered by the clean integration of Compellent storage with ESX hosts. With only minimal orientation to the Compellent terms and concepts, some very advanced storage/virtualization activities were accomplishable without ever leaving vSphere.

Data Agility and Mobility

The same Dynamic Block Architecture that enables blocks to seamlessly move between media tiers for optimized performance and recovery can also be used for online migration and synchronization of blocks between Storage Centers. This powerful capability simplifies the migration of virtual machines between sites and the migration of data between Compellent Storage Centers. The virtual machine migration capability is depicted in Figure 16.

In this example, servers located in two different data centers have been configured as a virtual server cluster. The virtual server platform is being used to move an application running in a virtual machine from one data center to the other. While the virtual machine is moved using industry standard virtual server technology (e.g., VMware VMotion, Microsoft Live Migration), Compellent Live Volume replicates changed blocks over the WAN and creates a single consistent view of data regardless of which data center is being used for production.

This powerful capability is an efficient, automated, and cost effective alternative to methods that have become available over the past few years. For example, VMware storage VMotion provides similar functionality, but requires too much server processing power and network bandwidth to make it practical for use between data centers. This can also be achieved with expensive clustering hardware that provides a view of storage between data centers deployed at each site.

ESG Lab Testing

ESG Lab tested the Live Volume feature by first creating an asynchronously replicated volume between two Storage Centers connected across a campus environment in a VMware environment.

Shown in Figure 17, ESG Lab selected a volume that was storing virtual machines from one of the hypervisors and was able to configure replication by simply right clicking on the primary volume and selecting which Storage Center would hold the alternate copy.

The most notable option in Figure 17 is the ability to automatically swap roles between the copies. This is the key benefit of the Live Volume feature in that, regardless of which host may be initially driving IO, the volume is adaptive. If the IO were to move from a host on one Storage Center to the other—for example, as part of a server-level recovery or migration scenario—the Storage Centers would mutually identify that the IO was now being generated from the other side and the direction of the replication traffic would transparently reverse. ESG Lab tested this by moving workloads between production hosts on either side of the Storage Center pair and observing the IO shifts between the sides and the directional shift of replication.

With Live Volume configured and replicating automatically, moving a running virtual machine to another data center was simply performed using the VMware vSphere client. Using the standard user interface for VMware VMotion, a virtual machine was dragged and dropped from one virtual server to the other. In the screenshot shown in Figure 19, the migrate option accessed from a right click on a virtual machine was used to launch the migration wizard. The virtual machine migration completed in less than five minutes. A heavy IO workload started before the VMotion was used to notice a momentary dip in performance, which lasted about three seconds. The virtual machine remained up and available throughout the test.

As seen in Figure 19, planned downtime or other re-hosting events was tested by ESG Lab with a few easy mouse clicks within vSphere that then leveraged Compellent’s Live Volume capability for a successful migration.

ESG Lab Validation Highlights

Along with the various “Why this Matters” conclusions throughout the report, the following is a summary of the actual hands-on tests and their outcomes:

• ESG Lab validated the performance and flexible cost-benefit of the Dynamic Block Architecture that underlies the Storage Center platform.
• ESG Lab confirmed that the right data blocks are placed on the most appropriate layer of storage based on relative demand and price/performance. In fact, Compellent anecdotally states that many customers often do not purchase any incremental tier-1 storage following the initial purchase because their capacity growth issues are addressed by the lower storage—not withstanding new workload requirements or performance innovations such as SSD as a new top tier.
• ESG Lab confirmed the ease of use and transparent operation of Compellent’s multi-tier storage solution.
• ESG Lab observed impressive integration between Compellent’s own management tools and those of hypervisors and server platforms as well as third-party monitoring, which collectively adds to Compellent’s value because the capabilities become attainable from  existing management interfaces.
• ESG Lab confirmed the Replay capabilities of Storage Center to yield site-level resilience, as well as easy volume-level data recoveries.
• Live volume was used to perform an online VMotion of a VMware virtual machine between data centers. The VMotion operation was easy to configure and monitor from the VMware vSphere console. The virtual machine remained up and available during the migration.

Issues to Consider

While ESG Lab found many commendable aspects of Dell’s Storage Center, the following are a summary of considerations for future development by the Compellent team and/or factors to consider by prospective customers:

• While the integration of Storage Center with VMware vCenter was impressive and logical, parity for other hypervisors is not yet in place. With the anticipated release of Microsoft’s next generation of management tools in System Center 2012, Compellent has an opportunity to provide similar integration into not only virtualization management (VMM 2012), but also systems management (OpsMgr 2012) which would likely provide parity for Citrix Xen configurations.
• While Live Volume can be used for online migration of storage between Compellent Storage Center solutions, it does not support data migrations between heterogeneous disk arrays from different vendors. If a downtime window cannot be tolerated, a traditional host-based data migration utility or a network resident data mover can be used to migrate storage from another product to a Dell Compellent Storage Center.
• While the VMware vSphere plug-in provides access to typical Storage Center management tasks, the ability to perform a Live Volume swap operation from a VMware management console is not yet supported. Dell advised ESG that this is planned for an upcoming release.

The Bigger Truth

One of the greatest challenges IT faces today is managing unabated data growth in dynamic, virtualized environments. Innovation and automation are needed to reign in the cost and complexity associated with unrelenting growth and ever-changing business requirements. Virtualization technologies are needed to simplify, consolidate, and automate routine IT functions.

Innovative virtualization technologies are being deployed by a growing number of IT organizations. A recent ESG survey of 1,602 IT professionals indicates that server virtualization is leading the charge. Seventy-four percent of respondents report that they are actively using server virtualization.[4] Server virtualization, like storage virtualization, is powerful game-changing technology that can be used to consolidate and simplify complex IT infrastructure. Like the early adopters of server virtualization, forward-looking IT managers are turning to highly virtualized storage solutions to reduce costs, increase efficiency, and enhance the recoverability and agility of a consolidated storage infrastructure. ESG has confirmed that Compellent Storage Center version 5.4 is a powerful virtual storage solution that’s built on top of a pool-based Dynamic Block Architecture.

ESG Lab tested several sought-after storage capabilities, including thin provisioning, remote replication, rapid disk-based recovery, virtual machine migration between sites, and deep integration with leading virtual server platforms. In all cases, the answer to “Why is this different from a traditional SAN solution?” came back to Compellent’s Dynamic Block Architecture. The Dynamic Block Architecture is the DNA that enables simple and capacity efficient deployment of new volumes on physical and virtual servers. Dynamic Block Architecture also makes it easy to move applications and virtual machines between sites from a storage perspective, and to quickly recover data with capacity and network efficiency in mind.

ESG Lab was impressed with the flexibility and power of the Dynamic Block Architecture when first testing Storage Center in 2008.[5] Valuable capabilities that are now known in the industry as thin provisioning and sub-LUN tiering were built into the architecture and have been proven in the field by thousands of customers since the first product was shipped in 2005. As legacy storage architectures evolve to support these valuable capabilities, the flexibility and power of the foundational Dynamic Block Architecture is turning into a competitive advantage for Dell.   Features like Live Volume, which provides cost-effective clustered access to a virtualized pool of storage over distance, would be very difficult to implement without a page-based architecture.

While advanced features like Live Volume are powerful, the bottom line with the Dynamic Block Architecture is its ability to reduce the overall cost of purchasing and maintaining storage. Writes are automatically directed to the highest performing tier. Data Progression moves infrequently used data to a more cost effective tier and Replays are stored on the most cost effective tier. Due to these core capabilities, the Dell team indicates that many customers are using affordable SATA drives for most of their capacity with a couple of fast SAS drives for an automatic performance boost. That’s significantly more affordable than a traditional disk array full of expensive FC drives.

ESG Lab commends the Compellent team for the forethought needed to envision the value of the Dynamic Block Architecture back in 2002. This architecture has created a sound foundation that makes it possible for Dell to cost effectively deliver features and value that established storage vendors are struggling to match. ESG Lab congratulates Dell for another smart acquisition that will continue to add value to its data center portfolio. At the end of the day, ESG Lab finds that the goal of developing efficient, agile, and resilient storage has been met with the latest release of Compellent Storage Center. IT managers shopping for their next storage infrastructure upgrade—especially those that have embraced the benefits of  server virtualization—should seriously consider the benefit of virtualizing their storage infrastructure with the growing family of Fluid Data solutions from Dell.

Appendix

Additional ESG Coverage of Compellent

ESG has covered (Dell) Compellent in previous reports, including:

• ESG Market Landscape Report, Replication Technologies for Business Continuity, August 2011
• ESG Market Landscape Report, Storage Tiering, July 2011
• ESG In the News, Dell nears Acquisition Deal for Compellent, December 2010
• ESG Technology Brief, Compellent – Extremely Efficient Storage, April 2009
• ESG Lab Validation, Storage Center 4.0, February 2008

[2] http://www.enterprisestrategygroup.com/using-esg-lab-workloads/

[3] Source: ESG Research Report, The Evolution of Server Virtualization, November 2010.

[4] Source: ESG Research Report, The Evolution of Server Virtualization, November 2010.

[5] A list of previously published Compellent-focused ESG publications can be found in the Appendix.

via HPC Roundup: Data Direct Networks, Virident, Intel, Ciena.

We can no longer continue to create proprietary monolithic “boxes” of functionality where complexity is layered upon complexity to deal with the complexity that was put there last year in what might be described as “Ponzi storage.”

What if we started over.  Seriously – if we did not have legacy storage architectures to deal with and we needed to build storage for a world built on virtual machines?  That’s what is intriguing (to me at least) about Tintri.  If we start from scratch we don’t need to worry about masking complexity with even more complexity.  If we started from scratch in a virtual world, would we even have to deal with mapping disks/LUNS/volumes/zones and worldwide names?   Or would we be dealing with datastores and virtual disks as a part of the VM environment, from a VM-vcentric view?

One quarter of the respondents we surveyed for our server virtualization report (published last November, requires log-in), reported server virtualization created a significant storage scalability problem, and 21% reported performance problems.  That’s just a couple of the storage issues–there are a number of challenges from needing faster provisioning to better training for the IT staff.   As long as we have proprietary, complex, monolithic boxes, we’ll continue to see challenges like this when trying to map an environment that was designed to be rather static–traditional storage arrays–to an environment designed to be virtual and dynamic–virtualized server farms.  Tintri tackles these challenges by taking a top-down approach to storage–from the VM view–rather than a disk-up view as the industry has designed most storage systems.  It adds some solid state storage and secret sauce and claims to attain some pretty significant performance numbers.  Most significantly, it eliminates all the complex layers that users have to map through to figure out their virtual machine storage environment.

It won’t be an easy road for Tintri.  As a start up, users will be wary about viability, longevity, and support.  All valid concerns.  But users need to balance the risk of going with a start up with the potential benefits.  Tintri, as a start up, is not trying to protect a legacy install base–unlike the established storage vendors that have a multi-billion dollar annual storage revenue stream to protect, Tintri can approach the storage challenge with a clean slate and do what is right for users.  That gives it a lot of flexibility in its approach.

The biggest challenges I see for Tintri are organizational and process oriented.  VM administrators will get this right away.  But storage administrators have an awful lot invested in existing storage environments, not just from a dollars standpoint but from a process standpoint.  Data protection, disaster recovery, archive, and tiering are often done at the LUN or file level.  And if we create a VM-based storage environment, who ends up managing storage?  The VM administrator or the storage administrator?  I believe the storage administrator role will be alive and well, and this will all land in his or her lap as IT continues on its transformation journey and over time offers VM-based storage services in addition to block and file-based storage services back to the business.  But storage administrators will have to get their heads around this and figure out the impact Tintri will have on other processes–but–and this is probably another whole blog–with Tintri, users wil be able to protect VMs, rather than LUNs.  It will tie much cleaner back into the application environment, and isn’t that what it’s all about?

You can read Terri’s other blog entries at IT Depends.

Background

ESG recently conducted a survey of 463 IT professionals in enterprise (i.e., 1,000 or more employees) and large midmarket (i.e., 500 to 999 employees) organizations. Respondents were asked to identify what they would consider to be important IT initiatives over the next 12-24 months. As one might expect from a captive audience of server virtualization users, there was significant focus on consolidation efforts (cited by 31% of respondents), creating an internal or “private cloud” infrastructure (24%), and increased IT automation (23%). Perhaps of greatest interest and significance, however, is that desktop virtualization (36%) ranked as the top IT initiative selected by respondents, as seen in Figure 1.[1]

The data center as we know it is in the midst of a significant transformation; the data center of tomorrow is virtualized. Server virtualization, while still maturing its overall capabilities, is here to stay. Desktop virtualization is on the same trajectory. Both are logical and compelling, operationally and financially, and both make business sense. IT is beginning to accelerate virtualization of its server and desktop infrastructure. The more advanced an organization in terms of virtualization deployments, the greater the level of benefit and value that organization can expect. With systems virtualization becoming ubiquitous, storage, at least storage as it has been deployed and used for decades, can be a significant obstacle that is slowing or stalling the successful and optimal advance of IT. And yet the tools exist to capably address the issue—with the virtualization of storage being a crucial element.[2]

DataCore Software

DataCore has been providing software-based storage virtualization for over 13 years with more than 6,000 customers and over 20,000 licenses deployed. DataCore’s mission is to enhance the value of the storage hardware users prefer to own.

Their software-based virtualization approach works across all popular brands and models of disk and storage arrays, providing an integrated and consistent set of provisioning, data protection and performance acceleration functions on an infrastructure-wide basis, regardless of device and without requiring the purchase of expensive add-on software for each array in an environment.

SANsymphony-V

DataCore SANsymphony-V is designed as a flexible software platform which runs on physical or virtual industry standard x86-64 Windows servers. The product is agnostic with regard to the underlying storage hardware and can effectively virtualize whatever storage is on a user’s floor, whether direct-attached or SAN-connected. Running on industry standard server hardware and Windows Server 2008 R2 enables SANsymphony-V to support a wide array of storage devices and SAN connectivity options, including iSCSI, Fibre Channel and Fibre Channel over Ethernet (FCoE) determined by the HBA’s or NICs installed in the DataCore nodes.

SANsymphony-V can be installed on dedicated servers to manage very large multi-petabyte, physical storage pools, or scaled down for smaller environments by co-residing in the same physical server hosting the virtual machines. Since SANsymphony-V is a native Windows application, it can be installed directly on Hyper-V, without having to live inside a virtual machine.

SANsymphony-V is built with business- and mission-critical applications in mind. The system is designed to enable users to provision, share, reconfigure, migrate, replicate, expand and upgrade storage without downtime or performance impact.

SANsymphony-V offers users a rich set of enterprise-class functionality, with workflow-oriented wizards to ease administration and management. High-speed caching, synchronous mirroring, snapshots, continuous data protection (CDP) rollback, online migration and upgrades, asynchronous replication with compression, (for bi-directional remote site recovery), and thin provisioning all can span heterogeneous storage because SANsymphony-V sits between the storage hardware and hosts. The goal of the balance of this Lab Validation report is to examine exactly how DataCore SANsymphony-V virtualizes storage and provides compelling business value to users.

ESG Lab Validation

ESG Lab performed hands-on evaluation and testing of SANsymphony-V R8 at DataCore’s Fort Lauderdale, Florida Headquarters. Testing was designed to demonstrate the ease of deployment and management of SANsymphony-V, while providing an efficient, highly available, fully virtualized storage environment. ESG also validated DataCore software’s ability, in conjunction with a hypervisor, to significantly reduce the cost and effort to gracefully transition from a physical to a virtual environment by repurposing industry standard servers and storage already in place.

Getting Started

Figure 3 illustrates the test bed used by ESG Lab for this Validation report. Six physical servers were utilized, all running Microsoft Windows Server 2008 R2. Four servers were configured as Hyper V R2 hosts and two servers were configured as SANsymphony-V R8 storage virtualization nodes.

The Hyper-V servers were attached via Gigabit Ethernet iSCSI SAN to the SANsymphony-V nodes, Fibre Channel is also available, as is any SAN protocol supported by the host. The SANsymphony-V nodes aggregated raw storage from the directly attached disks and presented shared virtual disks to the attached servers. The SANsymphony-V management console (which can be run local or remotely) provided a centralized management interface that was used to perform all configuration, management and monitoring tasks.

ESG Lab Testing

The SANsymphony-V Management Console, shown in Figure 4, presents a customizable view of the user‘s environment with areas on the left side of the screen that populate as the system is configured and resources are added. Users drill down into individual components and entities from high level categories to manage all aspects of their virtualized storage environment.

ESG Lab began with the Getting Started wizard, and followed the steps a new administrator would execute to virtualize and serve storage to hosts using SANsymphony-V.

First, a user ‘ITAdmin’ was registered and given full read-write privileges on the system. Next, the available SAN ports on the DataCore server (node) were assigned to specific roles, as shown in Figure 5. Front-end ports present virtual disks to hosts, back end ports talk to disks or storage arrays, while mirror ports are used for local (synchronous) mirroring between DataCore nodes.

The next step was to create a disk pool. In these tests, disk pools were built from direct-attached drives, but a disk pool could also contain LUNS on SAN-attached storage arrays. Figure 6 shows a disk pool being created from four 36 GB physical disks. Additional multi-terabyte pools were also created. Storage tiers could also be set up by creating disk pools with devices having different price/performance characteristics including basic SATA drives, SAS or FibreChannel drives, and Solid State Disks (SSDs).

Next ESG Lab registered a host, selecting a virtual machine on one of the four Hyper-v servers.

The final step was to serve a disk to the host. ESG Lab right-clicked on the host and selected ‘Quick Serve New virtual Disk’, as seen in Figure 7. The only parameter required was the capacity desired for the disk.

Thin provisioning, cache management, synchronous mirroring between DataCore nodes, as well as preferred and alternate paths to the virtual disk resources are all configured automatically by SANsymphony-V.

Figure 8 shows the new 1TB volume mounted by the virtual machine ‘Exchange-Host’. Note that the 1 TB disk is automatically thin provisioned and generated from a much smaller physical disk pool.

Fully Virtualized and Highly Efficient

DataCore SANsymphony-V is designed to reduce the number of servers and storage devices required in a given environment by intelligently and securely sharing heterogeneous resources among them, even when the disks and arrays are of dissimilar models and brands. ESG Lab’s goal in these tests was to demonstrate how SANsymphony-V can be used to transition from an underutilized physical environment to a flexible, fully virtualized environment using existing equipment and resources. Figure 9 shows the physical server environment before testing began: Six Windows Server 2008 systems, with a single application installed on each server and direct attached SATA hard drives.

Server virtualization software, like Microsoft Hyper-V, VMware vSphere, and Citrix XenServer, enables users to consolidate physical servers by moving applications into virtual machines, and allowing multiple applications to run on a single physical server. In order to leverage advanced capabilities like Live Migration or vMotion, the ability to move virtual machines between physical servers while running, requires that the physical servers have access to shared storage. Since, the direct attached storage inside each server is not shareable, customers who desire the benefits of virtual machine mobility must purchase new external SAN arrays.

DataCore software enables customers to share access to their direct-attached drives without having to replace them. DataCore passes through the original disk contents to the hosts once incorporated into the virtualized pool.

Figure 10 shows the test environment, fully virtualized. Note that the same six physical servers were redeployed into new roles and are now able to provide enterprise-class capabilities. Hyper-V enables all six applications to run on two clustered servers in the main datacenter, with one Hyper-V server running in the disaster recovery site as a standby for the production servers.

SANsymphony-V was installed on the remaining three servers, turning them into DataCore storage virtualization nodes. Although they are represented with different symbols in this diagram, those are the same physical servers that were hosting isolated applications shown in Figure 9. The original direct-attached physical disks were spread among the three DataCore storage virtualization nodes. The DataCore software makes the disk space shareable over a SAN.

The two synchronously mirrored DataCore nodes in the main data center provide high availability. They can be located up to 100 kilometers apart to split the main datacenter within a metropolitan area, avoiding disruption due to facility issues. Critical virtual disks are replicated asynchronously to the third node at the disaster recovery site to protect against regional disasters and outages. . This adds the flexibility to grow the entire storage environment, on the fly, using any storage.

In the following test, ESG Lab imported a physical File Server’s live data volume into SANsymphony-V and converted it to a fully virtualized, thin provisioned volume.

ESG Lab Testing

First, a virtual disk was created to act as a logical pass-through device for the imported file server volume. As seen in Figure 11, the virtual disk was created with mirroring disabled.

In the next step, ESG Lab selected the DataCore server and specified the file server’s data drive as a pass-through disk, as seen in Figure 12.

Figure 13 shows the virtual disk being served to a host (a virtual machine in the Hyper-V server), at which point the volume was brought online.

Once the volume was served to the host, its contents were mirrored to a virtual disk on the other DataCore server, as seen in Figure 14. The original disk drive was then fully virtualized in the background to take advantage of thin provisioning and other capabilities while its mirrored copy provided uninterrupted access. The original drive was then non-disruptively replaced.

The entire process, from start to finish, was accomplished with eight mouse clicks and no outages after the pass- through disk was served to the host.

Fast

DataCore uses advanced caching techniques to accelerate I/O response in much the same way as modern high-end storage subsystems do, with three significant differences:

• The caching software is infrastructure-wide, rather than confined to one device.
• Cache can reside on the same servers as the virtual machines.
• Server RAM is typically less expensive than dedicated storage array cache, so organizations can get more performance enhancement for their money.

DataCore’s cache operates across all disks in the storage pool including heterogeneous storage systems. It exploits the relatively low-cost memory in the commercial server platforms on which it runs, converting their RAM into high performance, scalable storage cache.

SANsymphony-V’s cache resides between the operating system on the hosts and the physical storage. Like other sophisticated caching techniques, it provides a variety of performance acceleration services including read-ahead (pre-fetch), write-behind (where IO is acknowledged as soon as it is written to redundant caches), and write-coalescing, which enables SANsymphony-V to organize the sequence of writes so that multiple writes can be executed in a single operation.

SANsymphony-V’s caching acceleration applies to all storage devices configured throughout a user’s SAN. The ability to allocate up to 1 TB of cache per node can significantly enhance the performance of both low-end storage devices and high-end storage systems.

ESG Lab Testing

ESG Lab used the Iometer workload generation tool to emulate a mix of real-world applications (Exchange, SQL Server, file server, media server), to simulate an I/O mix that’s typical of multi-user business productivity applications.[3] The objective was to determine the impact on performance when DataCore software is added into an existing hardware environment.

ESG lab ran a mix of four workloads on four individual physical servers with 2 internal SATA drives each under Windows Server 2008 R2. Next, the same workloads were run on four virtual machines in a single physical server running Hyper-v R2. The four virtual machines were running against a single disk pool made up of eight SATA drives, to keep the hardware identical between the physical and virtual environments.

Figure 15 shows the performance benefit SANsymphony-V delivered.

Table 1 lists the detailed results obtained in our lab testing.

What the Numbers Mean

• Performance improved significantly for every workload with the File Server workload posting the most significant improvement, more than 6x.
• The Exchange 2007 workload showed a 5x improvement. The physical server with DAS was able to support 322 heavy users, while the virtualized system showed the ability to support more than 1,500 users.[4]

Highly Available

DataCore takes business continuity very seriously and offers multiple, complementary methods to ensure that hosts’ virtual machine images and their associated data on disk remain accessible despite hardware faults, human error, and environmental disruptions. SANsymphony offers N+1 redundancy to eliminate single points of failure in a datacenter. A DataCore node and/or its storage resources can be taken out-of-service for repairs, upgrades, expansion and replacement without interrupting applications, and then put seamlessly back into full use.

This level of high availability is provided via synchronous mirroring between nodes and multiple I/O paths from hosts to nodes. DataCore recommends that customers place redundant nodes with their respective disk pools in separate rooms, ideally in separate buildings on a campus where a water leak or air conditioning problem, for example, can only disturb one of the nodes while the other transparently absorbs its load. Larger customers often operate distributed datacenters split between hot sites as far as 100 kilometers apart. .

Data layer protection is provided via snapshots (full clones or copy-on-first write differentials) as well as true Continuous Data Protection (CDP). Either can be enabled for individual virtual disks. When a volume is protected with CDP, SANsymphony-V logs and time stamps all activity to the virtual disk so that users can create a rollback at any point in time within the rollback window.

To guard against regional disasters, DataCore offers asynchronous remote replication over conventional LANs and WANs using industry standard TCP/IP protocols. SANsymphony-V automatically compresses the replication stream to reduce bandwidth requirements, allowing customers to use lower cost links with narrower bandwidth.

Figure 16 shows the ESG Lab test bed used to validate synchronous mirroring and asynchronous remote replication.

ESG Lab Testing

ESG Lab tested synchronous mirroring using a simulated Exchange server. The server was running in a virtual machine on a Hyper-V server in the main datacenter, and its database and log volumes were being served by SANsymphony-V via the East Node, seen in Figure 16. Iometer was used to generate Exchange database and log workloads against the two volumes.

ESG lab disabled the East Node while the workloads were running, and verified that the node was indeed down, as shown in Figure 17. Iometer continued to run without interruption, transparently failing over to the West Node.

ESG Lab restarted the East Node and verified through the SANsymphony-V management console that all disk access was still being serviced through the West Node, as shown in Figure 18.

After the SANsymphony log recovery process completed, the volume failed back to the East Node, again automatically and without interruption.

Next, ESG Lab enabled Continuous Data Protection on the disk ‘Exch Log’, highlighted in Figure 19. A batch file was run in a continuous loop, creating a series of small text files, one per second to simulate continuous writes to a log file.

Next, ESG Lab simulated an accidental file deletion event and stopped the batch file. To recover, a rollback point was created, at 2:12:02 PM, as seen in Figure 20.

Finally, ESG un-served the volume from the Exchange server, and reverted to the rollback. After serving the volume back to the Exchange server, the text files created with the batch file were confirmed to have been restored back to the 2:12:02 PM point-in-time of the rollback.

Next, ESG Lab tested asynchronous remote replication using a standard LAN connection and a WANem Wide Area Network simulator, which restricted bandwidth to a T1 equivalent (1.44 Mbps) and injected latency into the link, to simulate a real world connection between a main data center and a disaster recovery site 10 miles away.

A fixed 1 GB data set consisting of standard office files and digital photos was used to test replication. First, the robocopy command was used to copy the data set from a source volume on one side of the WAN Emulator to a target volume on the target side. As expected, the robocopy process took just under 2 hours to complete and delivered consistent throughput of approximately 1.4 Mbps.

ESG Lab copied the data set to a virtual disk on SANsymphony-V and configured asynchronous replication between the sites. First, a replication buffer was defined. This step is only required once at the source. Replication buffers allow asynchronous replication to occur over slow or intermittent links by buffering changes to a local disk. Next, the virtual disk to be replicated was identified, which was named Replication Source, as seen in Figure 21.

Figure 22 shows the local and remote DataCore servers being partnered in a replication group.

Next, the replication session was set up for the source volume, as seen in Figure 23.

Next, replication was started and the start time was noted. Replication completed in 1 hour and 12 minutes, representing a virtual throughput of 2.84 Mbps, or about double the raw speed of the T1 link.

After replication was complete, a host at the disaster recovery (DR) site was able to mount the remote copy of the virtual disk and access the data to take over processing duties. To simulate updates occurring at the DR site while the main datacenter is offline, new files were copied to the remote virtual disk. Before resuming normal business operations at the primary datacenter, the changes were automatically transmitted back to the main datacenter using SANsymphony-V Advanced Site Recovery. This feature effectively reversed the order of replication to expedite the switch back.

ESG Lab Validation Highlights

• DataCore SANsymphony-V R8 was very quick and easy to set up and manage. ESG Lab set up a two node, high-availability SANsymphony-V environment and was serving storage to virtual machines in minutes.
• With one click of a mouse, ESG Lab was able to serve a thin provisioned, performance tuned, fully protected, synchronously mirrored 1TB virtual disk to a Windows server in less than a minute.
• Migrating and importing disk drives from a working physical server into the virtual disk pool was a quick and easy process. The DataCore software provided a seamless transition from physical to virtual infrastructure while actually enhancing performance and availability.
• DataCore synchronous mirroring provided hosts with continuous access to virtual disks through a node outage with zero downtime and no interruptions to service.
• DataCore Continuous Data Protection (CDP) was easy to configure and use, enabling rollback to a specific point in time without having to create multiple snapshots.
• DataCore asynchronous remote replication was also easy to configure; compression and multi-streaming provided a 2x throughput enhancement over a simulated T1 link.

Issues to Consider

• While remote replication was easy to set up and use, initial replication of a large data set could take a very long time over relatively slow IP connections. The ability of a customer to ‘seed’ a replication target, using tape or removable disk, would enable much quicker movement of the initial data set to the remote site. As of this writing, DataCore has informed ESG Lab that data seeding is typically a procedure performed by the solution provider. DataCore informed ESG Lab that more automated means of data seeding will be available in a future release of SANsymphony-V.

The Bigger Truth

As server virtualization has matured, storage management has become more complex, and the dependency on centralized shared storage is changing the storage landscape. With applications detached from physical systems, their workloads have become more dynamic and unpredictable. These trends drive the urgency to virtualize storage. Virtualization is one of the few IT tools with a genuine ability to significantly address the challenge of unabated demand for a limited supply of the resources to deliver IT services and the tools to manage them.

Virtualization was a “nice-to-have” five or ten years ago but is now rapidly becoming a hard requirement for many production environments. IT has no choice but to virtualize. Much as the experience and success of server virtualization has transitioned from initially supporting test and development to now being a software infrastructure for data center architectures. Storage virtualization has to be a part of that virtual IT infrastructure.

SANsymphony-V R8 is DataCore’s newest release based on more than a decade of experience, and DataCore has thousands of users that will attest to its ability to deliver both quality and business value. The company now finds itself with incredibly relevant capabilities that truly matter to users. ESG lab tested DataCore SANsymphony-V and found the software easy to implement and to manage, virtualizing any storage infrastructure with enterprise class features and functionality while enhancing performance. DataCore SANsymphony-V kept data available and online through both planned and unplanned outages flawlessly.

The quest for realistic and affordable options to deal with the challenges inherent in IT virtualization and consolidation is daunting. Storage administrators are increasingly being retired and replaced or re-invented as network and virtualization administrators; storage management in a modern IT environment has to be simple and practical as well as functional as users will eventually be compelled to virtualize everything. The only variable will be timing; ESG Lab firmly believes that it would benefit any organization considering or implementing an IT virtualization project to take a long look at DataCore SANsymphony-V R8 storage virtualization software. .

It is robust, flexible, and responsive and it can deliver major value in terms of utilization, economics, improved response times, high availability (HA), and easy administration.

Appendix

[2] Source: ESG Market Report, The Future of Storage in a Virtualized Data Center, January 2011.

[3] Workload characterizations are listed in the Appendix.

[4] Exchange 2007 User profiles are listed in the Appendix.

[5] Source: ESG Research Report, The Evolution of Server Virtualization, November 2010 .

Background

ESG conducted an in-depth survey of senior IT professionals concerning their organizations’ IT spending plans and priorities for 2011. Survey participants represented midmarket (100 to 999 employees) and enterprise-class (1,000 employees or more) organizations in North America and Western Europe. The top ten responses are shown in Figure 1. Priorities were fairly evenly distributed with every listed priority getting at least one response. The top two priorities for organizations were increasing the use of server virtualization and managing data growth.[1]

In another recent study, ESG found that server virtualization is becoming ubiquitous in IT organizations of all sizes. Of 1,602 survey respondents, nearly three-quarters (74%) said their organization currently uses server virtualization. An additional 19% of organizations are in the evaluation or planning phase, leaving just 7% of the midmarket and enterprise organizations surveyed by ESG not currently using server virtualization and having no plans to do so.[2]

IT professionals who have experience with server virtualization know that like server virtualization, storage virtualization enables consolidation, which reduces complexity and cost. Storage and server virtualization are also similar in their ability to provide increased availability, fault tolerance, and mobility. Given the similarities, it’s not surprising that, in an earlier study, ESG learned that an increasing number of organizations are deploying server and
storage virtualization together. Fifty-seven percent had either already deployed storage virtualization in conjunction with server virtualization or planned on doing so within the next 24 months at the time of that survey.[3]

One of the most significant findings involved the substantial economic benefit of deploying an intelligent storage virtualization solution in conjunction with server virtualization. As an example, 79% of early adopters with a large SAN believe they have reduced their annual storage hardware spending to some degree, with a reported mean annual savings of 19.7%. Storage hardware savings were the most significant due to the ability to reclaim and re-use existing storage (the least expensive storage is the storage you already have). Consolidating storage software that previously ran on multiple servers or storage systems onto a centrally managed infrastructure also helped reduce storage software spending and storage administration costs.

IBM Storwize V7000

The IBM Storwize V7000 is a midrange disk system that has been designed to be easy to use and manage, enabling rapid deployment with minimal resources. Storwize V7000 is virtual storage that offers efficiency and flexibility through built-in SSD optimization and “thin provisioning” technologies while enabling users to virtualize and re-use existing disk systems, as shown in Figure 2.

Storwize V7000 advanced functionality also enables non-disruptive migration of data from existing storage, simplifying implementation and minimizing disruption to users.

IBM Storwize V7000 can be deployed in a data center to provide:

• Dramatic performance improvement with IBM System Storage Easy Tier functionality, providing automatic storage tiering between hard disk drives and solid state drives (SSD)
• Optimized capacity utilization and availability
• Significantly improved storage management ease-of-use
• A centralized platform for the management of SAN-attached storage capacity
• Heterogeneous interoperability between a wide variety of hosts, operating systems, and storage systems
• Online data migration services
• Remote replication and disaster avoidance services
• Space efficient provisioning and copy services

Storwize V7000 may be used to provide all these functions to enhance existing installed storage or as part of a new storage deployment.

The goal of ESG Lab’s testing of the IBM Storwize V7000 was to validate enterprise-class features and functionality including ease of use, non-disruptive virtualization, automated tiering, data mobility, and copy services. The performance benefits of IBM System Storage Easy Tier were also explored.

ESG Lab Validation

ESG Lab performed hands-on evaluation and testing of the IBM Storwize V7000 at an IBM facility located in Tucson, Arizona. A SAN-attached IBM 3850-X5 server was used during testing as shown in Figure 3 and documented in the Appendix. An IBM DS4800 storage array was used to validate the ability to virtualize and manage external storage.

Ease of Use

ESG Lab testing began with a high level examination of the Storwize V7000 user interface shown in Figure 4.

The Storwize V7000 graphical user interface is a browser-based, easy to navigate intuitive GUI. The home screen shown in this screen shot provides an excellent graphical flow chart of the system components and steps required to provision storage to host systems. The getting started view also provides e-learning and information center links that can be leveraged to provide detailed information on each component in the storage provisioning process.

ESG Lab Testing

ESG Lab leveraged the getting started view to confirm physical system components and resources as well as a gateway to configuring more advanced system features. The flow diagram and component detail information was repeatedly referenced during the configuration of many storage components including managed disks, pool, and volume creation to host provisioning and external device imports.

The Storwize V7000 GUI provides an intuitive view of the system components organized on the left side of each pane in column format, allowing the administrator to scroll from component to component with ease. Hovering over an individual component will provide drilldown options of the administration tasks that can be performed for that component. Figure 5 shows the administration options available for the Storwize V7000 volumes.

Figure 5 highlights the drilldown options for volume operations, but it should also be noted that the middle of the screen displays system status information in this view. The system status pane provides a high level detailed view of system information moving from left to right. Figure 5 displays a physical view of system capacity via the bar graph to the left and a component view in the middle of the pane. From the system status screen, users can hover over each object and display detailed system information. The blue section of the bar chart will display used capacity while the black section will show total system physical capacity. A click on the link below the physical display will present a table of system statistics to the right or users can click on a disk shelf to display detailed information on the drives in each enclosure.

The Storwize V7000 disk system presents storage as volumes. Volumes are virtual containers, abstracted from physical disk drives by the concepts of managed disks and pools presented to host servers. The volume by pool view allows the administrator to see detailed information for each Storwize V7000 volume, including its pool association. Figure 6 shows the volume details and administration options available for volumes in pool WEB-SP.

ESG Lab Testing

ESG Lab validated the ease of executing typical storage administration tasks with the Storwize V7000 by creating a thin provisioned volume copy in an existing storage pool. Figure 6 shows the volume copy creation steps as well as the percentage completed. Volume copies can be assigned to hosts and used for a number of operations (e.g., off-host backups, test/dev, or data migration).

Efficient Data Management

With the Storwize V7000, the storage administrator has the ability to deploy enterprise-class management features to improve storage efficiencies. Thin provisioning and snapshot technology are included in the Storwize V7000 code.

Thin provisioning employs an abstraction layer between the physical disk drive and the storage presented to the host system. With the Storwize V7000, physical disks are configured in RAID sets as managed disks, or MDisks. The MDisks are then assigned to storage pools. Virtualized volumes are created from pool storage and presented to hosts. The volumes can be created larger than the physical storage as shown in Figure 7. Because the volume is abstracted from the physical storage layer, capacity can be dynamically managed transparently to the host operating system as requirements change.

Figure 7 shows thin provisioned storage as compared to traditionally provisioned storage. The left side of the figure shows the one to one relationship of a traditional volume with a 5 TB volume presented to a host server consuming 5 TB of physical space. The right side of the figure shows a 5 TB virtual container being presented to a host server using only 100 GB of physical space.

ESG Lab Testing

ESG Lab tested thin provisioning on the Storwize V7000 by over-subscribing the storage array. Thin provisioned volumes were created with more capacity than the physical array as shown in Figure 8. The Storwize V7000 had a physical capacity of 38.8 TB and a used capacity 34.0 TB. ESG Lab created thin provisioned volumes to present a virtual capacity of 41.5 TB.

The bar graph in Figure 8 shows the used capacity in blue, the physical capacity in black, and the virtual capacity in green. At this point in ESG Lab testing, the Storwize V7000 had presented more capacity to host servers than it physically contained, yet the used capacity was only 34.0 TB.

FlashCopy is a copy service provided with the Storwize V7000 disk system. There are three different FlashCopy options available in the Storwize V7000—Snapshot, Clone, and Backup—as shown in Figure 9. Implementing FlashCopy enables the instantaneous copy of data from a source volume to a target volume.

The FlashCopy Snapshot is a thin provisioned copy with the ability to automatically expand as necessary. The FlashCopy Clone option is a copy with the same properties as the source volume. FlashCopy Backup is a copy with the same properties as the source and the ability to create incremental copies.

ESG Lab Testing

ESG Lab validated snapshot functionality on the Storwize V7000 by creating a FlashCopy Clone of an existing volume that was already presented to a Windows host with an 80 GB database file residing on it. Figure 10 show the first steps for creating a FlashCopy copy volume from the FlashCopy Actions menu.

The new Clone option was selected and mappings were created from source volume (MWL-JET1_01) to target volume (MWL-JET1) as shown in Figure 11.

When the Clone copy was fully synchronized on the Storwize V7000, the 80 GB database file was deleted on the Windows host. After the file deletion, the source and target mapping were reversed on the Storwize V7000. The Windows host system was rebooted and the 80 GB database file was immediately available on the host while the volume synchronization process ran in the background as shown in Figure 12.

Powerful Enterprise-class Features

The Storwize V7000 disk system is a feature-rich storage solution that provides a powerful set of management tools enabling great flexibility while helping to make the most challenging storage administration tasks easy to tackle. As shown previously in Figure 5, both internal and external storage can be virtualized and managed via the Storwize V7000 graphical user interface. The GUI provides an easy to use management console and includes configuration wizards for many administration tasks. The ability to combine multiple storage concepts like external storage device imports and thin provisioning add to the power of the Storwize V7000, creating flexible capacity management.

ESG Lab Testing

ESG Lab tested and confirmed the ability to virtualize external storage devices by importing, presenting, and managing a DS4800 volume and its associated data via the Storwize V7000 system. As shown in Figure 13, a DS4800 volume was imported to the Storwize V7000, combined with a thin provisioned internal volume, and presented to a host server with multiple virtual Windows machines.

The volume import procedure was conducted and configured using the Storwize V7000 Import Wizard. The wizard made the import procedure easy and intuitive by providing descriptions and configuration options during the short two-step setup process.

The Import Wizard is just one of the administration wizards that can be launched from the Storwize V7000 GUI. As shown in Figure 14, the Import Wizard was used to import a volume and its data from and external DS4800 storage array.

Figure 15 shows the pool selection screen—the second and final step of the volume import process.

The original volume data was maintained during the import process and once imported, presented back to the original Windows machine via the Storwize V7000.

Upon completion of the import process, a thin provisioned volume (Copy 1) was added to the original imported volume as shown in Figure 16.

Figure 16 shows the traditional provisioned Copy 0 volume and the thin provisioned Copy 1 volume. It should be noted that each volume contained the same 8.8 GB of used data. The traditional volume used 400 GB of physical space to store 8.8 GB while the thin volume only consumed the 8.8 GB of physical space to store the same data.

Predictably Scalable Performance

IBM uses the latest hardware and field-proven storage software from the SVC and DS8000 storage product lines to deliver the exceptional performance of the IBM Storwize V7000. The hardware platform is built for speed and scalability using a pair of highly available controllers powered by the latest Intel chipsets. Each controller is equipped with 8 GB of high speed cache memory. Up to eight Fibre Channel interfaces (8 Gbps) and four iSCSI interfaces (1 Gbps) are supported across a pair of controllers for host connectivity. A mix of SAS, Nearline SAS, and SSD drives is supported to meet a wide variety of price/performance requirements.

The performance of the hardware platform is amplified by field-proven IBM storage software including the SVC storage virtualization engine, the DS8000 RAID stack, and Easy Tier, illustrated in Figure 18, which automatically moves frequently-used data to high performance SSD drives and infrequently-used data to cost-effective Nearline SAS drives. Thousands of customers across the globe rely on the field-proven performance of the SVC engine at the heart of the Storwize V7000.

The performance scalability of the SVC engine has been demonstrated in published Storage Performance Council (SPC) results. SPC is a vendor-neutral standards body designed to be a source of comparative storage subsystem performance information. There are two application workloads supported by SPC: SPC-1 simulates an interactive database application and SPC-2 simulates a bandwidth-intensive application such as those that process large files, large database queries, or video on demand. When this report was published, IBM had the industry-leading SPC-1 result of 380,489 IOPS with a six-node SVC cluster in front of two IBM DS8700 disk arrays.[4] The SVC engine running inside a dual controller Storwize V7000 had excellent published results of 56,510 SPC-1 IOPS[5] and 3,123 SPC-2 MBPS.[6]

ESG Lab Mixed Workload Testing

ESG Lab has confirmed that the performance and scalability of the Storwize V7000 is well suited for a mix of applications running in a consolidated virtual server environment. The ESG Lab mixed workload performance benchmark was designed to measure the performance capabilities of a single Storwize V7000 storage system subjected to an IO-intensive mix of virtualized business applications running on a pair of IBM x3850 X5 servers in a virtual server environment powered by VMware vSphere. Taking a cue from the server-focused VMware VMmark benchmark, a cell concept was used during the design of this test. Each cell was composed of four applications (e-mail, database, web server, backup jobs), each running in its own virtual machine. IBM System x3850 X5 servers were used to exercise up to four cells and sixteen virtual applications in parallel. The Microsoft Jetstress utility was used to simulated Exchange 2010 traffic and the Oracle Orion utility was used to emulate OLTP and OLAP database traffic. The industry standard Iometer utility was used to emulate thousands of web server users and up to four backup jobs.

Storwize V7000 disk capacity was used for all storage capacity including VMware virtual disk files (VMDK), Windows Server 2008 R2 operating system images, application executables, and application data. The operating system images were installed on VMDK volumes. All of the application data volumes under test were configured as mapped raw LUNs (also known as raw device mapped, or RDM, volumes).

Mixed workload testing was performed with a single IBM Storwize V7000 equipped with 215 SAS drives. The Exchange VMs were configured with 64 SAS drives for the database and 25 drives for logs. Oracle was configured with 52 drives and the web server and backup workloads were configured with 32 drives each.

Volume ownership was balanced across the dual controllers within the Storwize V7000 and distributed evenly over the eight host interfaces. The volumes were spread evenly over two VMware host groups with a multipath policy of most recently used (MRU).

As shown in Figure 17, a single Storwize V7000 delivers the IO processing power and bandwidth needed to simultaneously support 54,208 simulated Exchange 2010 mailboxes and 5,015 Oracle Orion small database IOs per second and 849 MB/sec of throughput for large OLAP Oracle Orion operations and 5,015 simulated web server IOPs and 644 MB/sec of throughput for bandwidth-intensive backup jobs—all while delivering predictably fast response times. Note how the performance of the Storwize V7000 scaled well as a mix of real-world application workloads ran in parallel on up to 16 virtual machines running on a pair of powerful IBM x3850 X5 servers.

Easy Tier

The solid state drives (SSD) supported by the Storwize V7000 can be used to improve application performance while simultaneously reducing power and cooling requirements in the data center. Easy Tier increases the efficiency and simplicity of deploying SSD drives. Easy Tier, and its associated tool Storage Tier Advisor, help IT managers plan for, deploy, and manage the deployment of SSD drives in conjunction with traditional spinning hard drives.

As shown in Figure 18, Easy Tier uses sub-LUN data tiering technology that can be used to automatically move frequently-used data to high performance SSD drives and infrequently-used data to Nearline drives. Easy Tier can be configured to improve the performance for existing storage being virtualized by the IBM Storwize V7000 system and  can operate with any combination of internal or externally virtualized SSD or HDD devices.

ESG Lab evaluated the performance boost that can be achieved with SSD drives and Easy Tier. Twenty-four SSD drives were added to the pool of SAS drives used during the first round of mixed workload testing. An Easy Tier policy was configured to automatically move frequently-used application data to 24 high speed SSD drives.

After re-running the mixed workload on all of the available virtual machines for 24 hours, Easy Tier more than tripled the amount of mixed workloads the solution could handle. [7]

At the application level, Easy Tier delivered:

• 341% more web server IOs per second
• 277% more Oracle OLTP IOs per second
• 69% more Exchange 2010 users
• 33% faster Exchange database read response times
• 43% faster Oracle OLTP IO response times

ESG Lab Validation Highlights

• The Storwize V7000 was exceptionally easy to configure and manage. ESG Lab was able to provision virtualized, thin provisioned storage in less than 10 minutes using intuitive, automated wizards.
• ESG Lab performed a non-disruptive import of an external disk volume and presented the data intact back to the original host without requiring a reformat of the storage during import.
• IBM Storwize V7000 was able to thin provision the imported volume to better use capacity without destroying user data.
• An IBM Storwize V7000 and a pair of IBM x3850 X5 servers delivers the IO processing power and bandwidth to concurrently support up to:
  • 54,208 mailboxes using the Microsoft Exchange 2010 Jetstress  utility
  • and 5,015 database IOs per second for small OLTP IOs with the Oracle Orion utility
  • and 849 MB/sec of throughput for large OLAP Oracle Orion operations
  • and 5,015 simulated web server IOPs
  • and 644 MB/sec of throughput for simulated backup jobs
  • with predictably fast response times and scalability
• As the number of virtual machines sharing a single Storwize V7000 increased, performance scaled in a near linear fashion with predictably fast response times (5.2 to 11.3 milliseconds for e-mail database reads, 4.72 to 5.06 milliseconds for Oracle OLTP IO operations).
• Easy Tier and 24 SSD drives more than tripled the mixed IO capacity of the Storwize V7000 (3.21 times more) as it noticeably improved application-level performance:
  • 33% faster Exchange response times
  • 43% faster Oracle OLTP database response times

Issues to Consider

• ESG Lab tested a Storwize V7000 system with a total of 240 drives (216 SAS, 24 SSD), which was beyond the maximum number of supported internal drives when this report was published. In this context, internal drives are the drives located within drive expansion trays connected to, and managed by, Storwize V7000 controllers. This limitation does not apply to the drives within a disk array that is externally virtualized with the Storwize V7000. IBM has publically stated that support for 240 internal drives is planned for general availability in early 2011.
• The field-proven architecture of the SVC currently support SVC code running on up to four pairs of controllers managed as a single system. The IBM Storwize V7000, which leverages the same SVC code base, currently supports only a single pair of controllers. ESG Lab believes that support for more than a single pair of controllers would be a welcome future enhancement for organizations that would like to scale the benefits of a singly managed pool of V7000 storage.
• ESG Lab is confident that most IT organizations will exceed the Easy Tier and SSD benefits documented in this report. This is due to the fact that real-world applications tend to have concentrated hot spots that are great candidates for Easy Tier and SSD. In contrast, the synthetic benchmarks used during ESG Lab testing are more uniformly random than real-world applications. As a result, ESG Lab believes that less SSD capacity should provide similar—if not better— performance benefits in real-world production environments.
• The test results/data presented in this document are based on industry-standard benchmarks deployed in a controlled environment. Due to the many variables in each production data center environment, it is still important to perform capacity planning and testing in your own environment to validate a storage system configuration.

The Bigger Truth

One of the largest challenges facing IT today is managing unabated data growth in dynamic, virtualized environments. Innovation and automation are needed to reign in the cost and complexity associated with unrelenting growth and ever-changing business requirements. Virtualization technologies are needed to simplify, consolidate, and automate routine IT functions.

Innovative virtualization technologies are being deployed by a growing number of IT organizations. A recent ESG survey of 1,602 IT professionals indicates that server virtualization is leading the charge. Seventy-four percent of respondents report that they are actively using server virtualization.[8] Server virtualization, like storage virtualization, is powerful game-changing technology that can be used to consolidate and simplify complex IT infrastructure. While the cost savings of consolidation drove the first wave of server virtualization adoption, the second wave is being driven by the mobility, flexibility, and enhanced fault tolerance that can be achieved with a virtualized IT infrastructure.

Like the early adopters of server virtualization, early adopters of storage virtualization report that management complexity has been simplified and costs have been reduced. Given the synergistic benefits, it is not surprising that ESG research indicates that a growing number of forward-looking organizations are deploying storage and server virtualization together. Fifty-seven percent have deployed both already or plan on doing so.[9]

ESG has confirmed that the IBM Storwize V7000 is a modular storage system with built-in efficiency and storage virtualization. Presenting virtual disk capacity on a just in time basis, space-efficient virtual disks eliminate the cost of unused and unallocated storage as they simplify the task of assigning storage capacity to applications.  Leveraging the field-proven storage virtualization capabilities of the IBM SVC product line, the Storwize V7000 provides a centralized platform for valuable data services including online data migration, copy services, mirroring, and remote replication.

ESG Lab was particularly impressed with the exceptional ease of use and performance of the IBM Storwize V7000. The management GUI, which was derived from the IBM XIV storage product line, is wizard-driven and extremely easy to use. Industry standard SPC results and ESG Lab mixed workload testing in a VMware environment confirmed the performance prowess of the platform. Performance testing with Easy Tier and SSD drives reduced application-level response times as it more than tripled the amount of work that the virtualized infrastructure could handle.

IBM Storwize V7000 is an easy to use, rock-solid midrange storage platform with sophisticated enterprise-class functionality and built-in storage efficiency. ESG Lab believes that IT managers within mid- to large-sized organizations shopping for their next storage infrastructure upgrade—especially those with legacy storage they’d like to continue to utilize—should seriously consider the IBM Storwize V7000.

Appendix

[2] Source: ESG Research Report, The Evolution of Server Virtualization, November 2010.

[3] Source: ESG Research Report, The Impact of Server Virtualization on Storage, December 2007, N=332.

[4] http://www.storageperformance.org/benchmark_results_files/SPC-1/IBM/A00087_IBM_DS8700_SVC-5.1-6node/a00087_IBM_DS8700_SVC5.1-6node_executive-summary-r1.pdf

[5] http://www.storageperformance.org/benchmark_results_files/SPC-1/IBM/A00097_IBM_Storwize-V7000/a00097_IBM_Storwize-V7000_2-node_SPC1_executive-summary.pdf

[6] http://www.storageperformance.org/benchmark_results_files/SPC-2/IBM_SPC-2/B00052_IBM_Storwize-V7000/b00052_IBM_Storwize-V7000_SPC2_executive-summary.pdf

[7] For more detail on the ESG Lab mixed workload testing of the IBM Storwize V7000, see the full ESG Lab Validation Report, IBM Storwize V7000: Real-world Mixed Workload Performance in VMware Environments, January 2011.

[8] Source: ESG Research Report, The Evolution of Server Virtualization, November 2010.

[9] Source: ESG Research Report, The Impact of Server Virtualization on Storage, December 2007, N=332.

Take, for example, the storage domain. What did the advanced users report as the biggest impact to their storage environment? Far and away the number one response (52%) was increased use of SAN-based storage. Next, they reported having to change their disaster recovery strategy (38%) and increase their use of storage virtualization technology (33%). Other interesting takeaways include 20% reporting they needed to get additional training and certification for storage administrators. When we polled the advanced users regarding what they believed they would need to enable more widespread use of server virtualization, the top answers were faster storage provisioning and additional training for IT staff (tied at 35%). Good information to know when you are just starting out.

Fortunately, in addition to the storage teams, we also surveyed the application, server, networking, and security teams to gather this information. If you are interested in learning more, we will be sharing this data, along with more examples from your peers, at our ESG Ahead of the Curve event on April 14th in Boston, MA. For more information, check out this site:

www.esg-ahead.com

I look forward to seeing you there.

You can read Bob’s other blog entries at Data Center Continuum.