The idea that virtualization technologies are revolutionizing the market is no longer novel. Though the conversation has shifted toward cloud computing, server virtualization remains the foundation. Driven by numerous benefits including lower capital and operational costs, greater IT efficiency and agility, and improved application performance, server virtualization is being deployed rapidly and near-universally by large midmarket and enterprise organizations. These benefits, realized by neophyte adopters, are merely the tip of the proverbial iceberg; organizations with more extensive deployments have realized improvements in application provisioning, automation of IT tasks, application availability, disaster recovery, business continuity, and application development.

The acceleration of server virtualization investments and cloud computing initiatives, combined with the potential change in IT infrastructure consumption, is rapidly transitioning the IT conversation away from a technology focus to more of a business-oriented approach. Today, the discussion is more about operational efficiency and economics. While the initial success of server virtualization was tied to the reduction of capital expenditures and containment of IT resources, the next phase is poised to focus more on its operational benefits and less and less on the technical nuances of one solution over the other. Procuring virtual computing infrastructures is going to become more of a business decision than most IT organizations are accustomed to. The buyers of virtual computing infrastructures are not concerned with the “speeds and feeds” of the infrastructure; they are focused on business priorities, time to market (cycle time), and ultimately improving IT service levels to the businesses they support. Server virtualization and cloud computing are acting as the ideal catalysts to introduce this strategy to mainstream IT and change the way businesses consume IT infrastructure.

IT is at a tipping point as more organizations consider investing in this new consumption model. This report will help to identify:

• The key benefits and driving factors of virtual computing infrastructures
• The primary challenges and concerns surrounding virtual computing infrastructures
• Tips for IT to use to help decide when to make a shift to a virtual computing infrastructure
• Approaches for  making an educated investment that fulfills short-term growth requirements and accommodates long-term IT strategies
• Potential vendor solutions

Though the report will discuss the deployment options that best suit future cloud initiatives, it will focus on the foundational infrastructure for x86-based server virtualization, not on the managerial add-ons and solutions that enable organizations to achieve the primary tenets of cloud computing. As a result, while some vendor solutions incorporate one more of these tenets, the solutions themselves center on virtual computing infrastructures.

Meantime, it’s very normal to make predictions at this time of year, with blogs and articles awash with prognostications…I won’t be immune from that over the coming weeks but for now I figure I’ll do something a little different by looking backwards in order to look forwards; and in so doing come up with one prediction of which I’m supremely confident.

Of course, once I found the stash of old pocket diaries I couldn’t help but look through them, whether or not they were applicable to the naturalization paperwork. Apparently I used to have a life! But I figure you’re not that interested in the plays I saw, parties I attended, and parenting events I undertook!? So I picked a few things of interest…before wondering what we would have predicted for 2012 from a decade or two ago?

First off, some of you may be too young to really have appreciated a good pocket diary – these things were mines of information….from world time-zones and public holidays (pretty useful) to wine vintages, metric/imperial conversions and atlases so tiny that you could barely make out the continents let alone cities. But there were also pages to note your expenses (no Excel back then!) and I was able to look back at all my achieved sales numbers through the years – at least until everything went electronic and of course I have no idea where, or if, the more recent records are! [I'm sure there's a pertinent point in that!] Of rather more drama, it reminded me of one of my managers in the UK that was murdered (the rumor was that it was a ‘hit,’ although whether the result of marital issues or conducting ‘dodgy’ leasing deals was never clear). Even more dramatic is the fact that the map of the World has changed notably in the last few decades….Europe alone looks crucially different from when I was selling round tape drives to British Airways! On an almost similar theme (because Germany is one of those changed geographies), but of no marked importance to the world, I found a distressingly large notation of when I got my first BMW….and the polo events that they’d invite me to, not realizing that I was a meager company-car-3-series-driver and not likely to be splurging big, real money of my own anytime!

So, trying to look forwards as if from back then, what can I say? I could lie and tell you that today’s technologies were clear to all, but that would be disingenuous in the extreme. After all – just to put it in perspective – I’m writing this blog on my BlackBerry in an airport, and later I’ll – wirelessly of course – clean it up and load it to the ‘Cloud’ site that supports it. Heck, I thought I was cool and space-age when I first got a car-phone, so I can’t lay claim to having figured out the impact or capabilities of this ‘Internet Age’! Actually maybe it was/is the ‘Information Age’….I know this for a fact because I can see that one of the myriad courses I attended back then was at Cranfield Management Centre in 1997 and it was called “Reinventing Competitive Advantage in the Information Age!”

Computing in the years before that – at least to me as an early business user – was lots of columns of green numbers on a black screen…..there were no products with an ‘i’ in front of them back then, and the line between corporate and personal computing (to the limited extent it even existed) was very solid and very wide. Basically, there were mainframes doing big, important work and people in corduroys and white coats that operated them. In fact, that perhaps brings us to one of the big changes – and it’s not technical at all. As we’re re-creating the concepts of a mainframe world in many respects (VDI running in the things called Clouds, over the Internet, accessing Big Data and interconnected up the wazzoo, create a massive global mainframe ‘feel’) the attire of choice for IT folks is getting a big change – and it’s not even jeans, as these days our web-based world means PJ’s are just as likely to be the sartorial choice.

What then is my semi-serious, but likely correct, prediction for 2012 and beyond? Forget the evolutionary technical advances, and watch the advance of the flannel revolution! This may read like just a whimsical point, but – like my old hand-written pocket calendars – there’s a lot more in it than you’d think.

You can read Mark’s other blog entries at The Business of Storage.

Indeed, more organizations are looking to deploy desktop virtualization solutions to deliver a consistent image, complete with all applicable business applications to any device an employee chooses to leverage in order to solve this problem. It enables IT to provide greater levels of flexibility to its employees and yet still maintain control over the data, especially sensitive data, which will still be housed in the centralized data center. However, with most pilots done in a limited data center environment, organizations need to be more aware of the impact this may have on the network, especially if they are planning on deploying it in any remote or branch offices. Like any centralized application, if not done correctly, poor performance and availability issues will lead to user frustration and eventually abandonment, potentially wasting a great deal of money (think about some of the problems with the initial centralized SharePoint deployments, etc.).

In 2012 this may be even more of an issue, as ESG research indicates that, while 25% of respondents to our Remote Office/Branch Office Technology Trends survey indicated that they have already deployed centralized virtual desktop solutions at ROBOs, another 42 % plan to do so over the next 18 months. So now is the time to ensure that your virtual desktop initiative will not be hindered by the WAN. With the proper insight (network management tools) and testing (make sure the POC includes at least one typical ROBO location), problems can be averted. In some cases the WAN link you already have may be adequate, in others, you may need more. In those cases, I would recommend investigating optimizing the WAN link before simply ordering more bandwidth as over time it could save a significant amount from your operational budget. It should be noted that while WAN optimization technology wasn’t originally designed to optimize virtual desktops, many vendors have recognized the need and are rapidly adding support for different desktop virtualization vendors and their specific technologies. The key will be knowing which WAN links need additional support and then matching the appropriate WAN optimization vendors to your desktop virtualization solutions in advance of the formal rollout in order to increase your chances for success.

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

I can’t even begin to fathom how much productivity time is lost or abandoned altogether simply due to how long it takes a system to “fire up.” I get that I may want to warm up my car on a cold morning, but the majority of times I like to get in it, buckle up, turn the key, and go. We don’t have to “fire up” our smartphones–they just work and provide instant access to calling and applications. It baffles me when I watch how long it takes some people to boot their systems or simply just not turn it on at all because of how long it takes to log on. The scary part is that IT is going to get more and more pressure from its users. I shudder to think how much productivity would be gained in any company if all endpoint devices instantly turned on and provided instant productivity. I suspect the impact would be dramatic.

I believe this is one of the many reasons we are starting to see such an insurgence of smartphone and tablet usage inside companies – they work – instantly! Laptops and desktops need to do the same. My informal survey of MAC users clearly validates the fact that one of the best benefits of the MAC is its ability instantly turn on, which alone is worth the price of admission. I’m positive someone can show me a Windows machine that does the same, but the majority of Windows users I see have to “fire up” their system.

Solid-state drives can help, but feel like more of a band-aid than a true solution. I think desktop virtualization has some very interesting technology that can help solve the “fire up” use case and may be reason enough to give it a hard look and factor into the investment.

Users are naturally going to migrate to the path of least resistance and we are already seeing this with the usage of smartphones and tablets as mentioned above, but is IT ready to support the proliferation of these devices? Have they considered the impact to the way they currently deploy and maintain applications? Have they thought through security challenges? I suspect in 2012 this will be a hot item for business to address. We are already seeing some leaders in IT stand out ahead of the curve, but for the most part, we are just getting started on a major transformation in the way we design, develop, deploy, maintain, consume, and pay for applications. This is very exciting and a tremendous opportunity for IT technology innovators!

You can read Mark’s other blog entries at Liquefying IT.

At a high level, the research identified these trends and opportunities:

• Forecast for greater use of SaaS/Cloud
• Remote/mobile user support increasing
• Desktop virtualization deployments expected
• WAN optimization important

Specific to desktop virtualization, ESG discovered that 25% of respondents currently leverage a desktop virtualization solution and an additional 42% plan to in the next 12 to 24 months, while 31% report no plans.

It should come as no surprise that desktop virtualization is gaining momentum in ROBOs. The question that arises is how do the additional 42% plan to adopt desktop virtualization solutions in the next 12-24 months. Market players focused on desktop virtualization solutions should consider specific solutions, product offerings, and go to market campaigns to specifically address this significant market opportunity.

You can read Mark’s other blog entries at Liquefying IT.

In the future, I anticipate that there will be more business specific applications leveraged and organizations will each have their own app stores to enable even more productivity from mobile devices and remote workers. Think of a virtual desktop that contains all relevant business apps delivered to a mobile device. However, the key to driving adoption will be to ensure that the end-user “experience” is a positive one. We have all had the frustration of trying to pull up a site or app that is loading or responding slowly. For me that prompts a quick thumb strike to the button on the bottom of my iPhone to cancel the app.

Businesses need to be more aware of the user experience when working from a mobile device or from a remote location. Most organizations have taken steps to optimize connections between data centers and to remote offices, but there is still more to be done to guarantee optimal performance to mobile devices and remote workers. Many WAN optimization companies are now extending those technologies to include PCs, MACs, and mobile devices based on Android and IOS. With ESG research indicating that the number of remote workers will be steadily increasing over next few years, it may be well worth an organization’s time to explore options for optimizing the connectivity for remote workers and mobile devices.

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

Introduction

Background

In recent years there has been a significant increase in the number of organizations considering flash-based solid-state drive (SSD) solutions for data center storage.  When ESG asked IT managers their reasons for assessing this option, improved performance topped the list, with 83% of respondents citing it as a consideration and 31% citing it as the most important consideration (see Figure 1). A significant majority of those surveyed also mentioned improved reliability (69%), improved power and cooling efficiency (65%), and the ability to consolidate disk drives (57%). While no single consideration outweighed the others as the most important reason, performance and reliability were mentioned as most important by 60% of respondents.[1]

While solid state drives have been used in consumer products such as MP3 players for a decade or more, data center deployments are new in the last couple of years. SSD implementations have provided organizations with orders of magnitude improvements in IO performance compared to traditional hard disk drives (HDD). A single SSD can deliver performance that might take hundreds of HDDs as it eliminates storage bottlenecks for applications that require fast performance and low latency. Today’s SSDs are also highly reliable with a life expectancy around ten years, and are much more power efficient than mechanical, spinning HDDs. These features make SSD attractive from an operational perspective, but the high cost of most SSD solutions compared to traditional disk arrays is making the investment hard to justify for most organizations.

Introducing the WhipTail XLR8r

The WhipTail XLR8r is an SSD appliance that supports up to 12 TB of MLC flash in an industry standard rack-mounted 2U form factor (see Figure 2). With an entry-level price of $49,000 and a price per gigabyte that’s comparable with a traditional enterprise-class disk array, the XLR8r appliance is designed to cost effectively meet the demanding performance needs of virtual desktop infrastructure deployments. The XLR8r family of appliances is also well-suited to meet the needs of a variety of performance-intensive application workloads including business-critical multi-user databases, e-mail, and high performance computing (e.g., financial analytics).

The patent-pending design of the WhipTail Racerunner OS (RROS) is designed to overcome the write performance and longevity challenges associated with cost effective MLC flash technology. The XLR8r appliance performs random writes faster than random reads, a unique and valuable capability given the fact that MLC flash components write slower than they read. This capability is ideally suited for a virtual desktop infrastructure as multiple users sharing a common desktop image flips IO from mostly random reads for a single user to mostly random writes for a mix of users.

XLR8r appliances support a variety of block-based host connectivity options including iSCSI (1 or 10 Gbps), Fibre Channel (2, 4, or 8 Gbps) and Infiniband (40 Gbps). The NFS and CIFS file-based protocols are supported as well.   The XLR8r has a simple user interface that’s designed to make it easy to deploy a turbocharged storage infrastructure that works seamlessly with existing applications.

The balance of this report explores the benefits of WhipTail XLR8r solutions including:

• Cost effective efficiency
• Write optimized high performance
• LUN and done simplicity
• Durable longevity and high availability

ESG Lab Validation

ESG performed hands-on testing at WhipTail Tech’s corporate headquarters in Whippany, New Jersey. Hands-on testing with a VMware virtual desktop infrastructure and industry standard workload generators was used to validate the high performance, longevity, fault tolerance, and simplicity of the XLR8r.

High Performance

ESG Lab testing began with a VMware View environment designed to simulate 300 users performing typical activities on virtualized desktops running the Microsoft Windows 7 operating system. As shown in Figure 3, four physical servers were used to host the virtualized desktops. The servers were connected via 10 Gbps Ethernet iSCSI to a WhipTail XLR8r with 2.2 TB of usable capacity spread over 12 MLC flash drives.[2]

VMware View linked clone technology was used during ESG Lab testing. This technology is often used in VDI solutions to reduce the storage capacity needed to support a large number of consolidated virtual desktops.  Instead of storing and maintaining a disk image for each user (e.g., the Microsoft Windows 7 C: drives during ESG Lab testing), a single image is used to store a shared “gold” image of operating system data. Updates that are unique for each user are saved and tracked. This creates a write- intensive IO workload that is a challenge for traditional disk arrays and not optimized for most SSD solutions, which tend to write slower than they read.

AutoIT, a freeware automation language for Microsoft Windows, was used to emulate users performing typical Windows 7 desktop activities:

• Opening and editing Microsoft Word documents
• Opening and editing PowerPoint presentations
• Reading PDF documents
• Viewing a video

A boot storm was tested as well. Boot storms often occur at the beginning of an employee shift and occasionally after planned or unplanned maintenance. It is essentially a storm of performance-intensive IO activity that’s not well-suited for traditional disk array architectures The VMware vSphere Client management interface was used to monitor VDI performance. IO throughput for one of the four physical VDI servers is shown in Figure 4.

Four physical servers are shown toward the top left with a subset of the 300 virtual desktops listed below. The graph shows the IO throughput for the first physical server in stacked form with the IO throughput of each virtual desktop shown as a separate color. Similar tools were used to monitor IO performance and latency as 300 simulated VDI users and a boot storm were tested. The results are summarized in Table 1.

What the Numbers Mean

• The VDI workload generated an average of 23 IOs per second per desktop. This level of activity is slightly more aggressive than the 20 IOs per second measured by ESG Lab with the Microsoft Performance Monitoring utility for a knowledge worker with heavy IO usage over multiple eight-hour shifts.
• The boot storm was extremely IO-intensive as it consumed up to 900 MB/sec of aggregate throughput.
• The low write latency of 2.2 milliseconds recorded during VDI workload and boot storm testing is very good, especially when compared to the performance of traditional disk arrays.

Host resources were the bottleneck during the first round of VDI performance testing with AutoIT. With a goal of determining the maximum number of virtual desktops that can be supported, the industry standard open source IOmeter utility was used to emulate the storage activity of a larger number of users. An IOmeter VDI workload developed by ESG Lab was used during this phase of testing.[3] The number of IOs per second reported by the IOmeter utility was used to calculate the theoretical maximum number of desktops that can be supported by the XLR8r appliance that was tested. The results are summarized in Figure 5

What the Numbers Mean

• A single XLR8r delivered 87,107 IOs per second for a VDI workload during IOmeter testing.
• At 23 IOPS per desktop, a sustained IO rate of 87,107 IOPS can support up to 3,787 heavy VDI users.
• A traditional disk array would require between 348 and 871 hard drives to sustain this level of performance.[4]
• A single XLR8r consuming only two units of rack space and drawing less than 200 Watts of power delivered the performance of multiple racks full of power hungry disk drives.

Next, ESG Lab performed a series of tests with the ATTO Disk Benchmark and IOmeter utilities in order to understand the underlying performance characteristics of the WhipTail architecture, including its ability to accelerate VDI performance with random writes that perform faster than random reads. Four physical servers, one virtual server, and the iSCSI protocols were used during this phase of testing. A high performance 40 Gbps Infiniband connection was tested with a goal of measuring the maximum throughput and minimum response time capabilities of the XLR8r.

The results of the ATTO Disk Benchmark test with a single volume accessed over a 10 Gbps iSCSI connection are shown in Figure 6. Note how write performance, shown in red, is better than read performance, shown in green.

IOmeter Performance characterization results are summarized in Table 2  and Figure 7.

What the Numbers Mean

• The WhipTail architecture performs random writes faster than random reads, which is ideally suited to meet the demanding IO requirements of a VDI environment.
• A peak throughput of 2 GB/sec over a 40 Gbps Infiniband connection is extremely high. High throughput not only increases productivity during VDI boot storms, it also accelerates the completion of bandwidth-intensive applications running on a high speed SSD solution (e.g., financial analytics, real-time data capture, video editing).
• A response time of 0.12 milliseconds for a single stream of random 4 KB writes is the fastest IO response time that ESG Lab has witnessed over more than eight years and dozens of storage system validation projects.
• The sub-millisecond single stream response times of the XLR8r are 10 to 100 times faster than a traditional disk array.
• Extremely fast response time improves user productivity as it accelerates applications responsiveness and increases the number of users that can be supported on a consolidated storage infrastructure.

Longevity

A common concern about flash is its longevity. This is due to the fact that flash cells must be erased before they are updated when an application performs a write operation. Flash components are often rated for one million write/erase cycles, but beyond several hundred thousand, the data retention period can drop to just a few weeks.  Because of their limited lifetimes, flash cells must be stressed uniformly so that hot cells don’t cause premature device failure. This is done through a process known as wear leveling. Just as disk drives keep a pool of spare blocks for bad-block remapping, flash devices reserve a pool of spare blocks for wear leveling. A pool of pre-erased blocks is typically maintained as well for performance.

Two types of flash are used in enterprise-class storage solutions: Single Level Cell (SLC) and Multi Level Cell (MLC).  SLC tends to have better longevity, but is less dense and costs more than MLC. Enterprise-class storage systems from major vendors have been supporting SLC flash drives for the past few years. WhipTail’s OS has been designed from the ground up to take advantage of the lower cost and higher density of MLC flash while delivering enterprise-class levels of longevity, performance, and durability.

As shown in Figure 8, the XLR8r extends longevity with patent pending algorithms that are layered on top of the continuously improving wear leveling technology in commodity MLC flash drives.  Longevity is delivered using a combination of proactive monitoring, wear leveling, load balancing and wide striping over groups of MLC flash drives in the system.

LUN and Done Simplicity

The XLR8r works like a standard disk array, presenting disk capacity as a standard block-based device. ESG Lab tested the LUN and Done simplicity of configuring a new XLR8r device for use in VDI infrastructure.  The steps and elapsed time required to power on and configure a new LUN are summarized in Table 3. A screenshot of the intuitive menu driven configuration process is shown in Figure 9.

High Availability

The highly available architecture of the XLR8r uses advanced RAID algorithms and redundant components to ensure continuous operation in the unlikely event of an MLC flash drive failure, a power supply failure, or a failed host interface.

ESG Lab injected errors while a heavy VDI workload and video were being serviced by an XLR8r. As shown in Figure 10, a redundant iSCSI host interface was disconnected and an MLC flash drive was removed. A redundant power cable was unplugged as well.

The iSCSI multipath IO (MPIO) driver that’s built into the Microsoft Windows Server operating system provided transparent failover when an iSCSI cable was disconnected. The XLR8r RAID algorithms ensured that IO processing continued without interruption when an MLC drive was unplugged.

Application workloads continued without interruption during ESG Lab error injection testing. Error messages and system status were monitored via the WhipTail management GUI. The ability to configure e-mail notification and SNMP traps for automated alerting was noted.

ESG Lab Validation Highlights

• VMware View VDI testing with 300 virtual desktops confirmed that the XLR8r can be used to cost effectively alleviate VDI performance bottlenecks.
• Excellent write response times of 2.2 milliseconds and a peak throughput of 900MB/sec were observed during a 300-user VDI and boot storm test.
• The ATTO Disk Benchmark and IOmeter utilities confirmed that the XLR8 writes faster than it reads, a capability that’s ideally suited for write-intensive VDI workloads.
• A peak throughput of 2.1 GB/sec and an extremely fast write response time of 0.12 milliseconds were recorded with a 40 Gbps Infiniband host connection.
• Write amplification and write usage statistics were used to confirm that a heavily-used system in the WhipTail lab has a projected longevity of 12 years and a well-used system in the field is expected to last 22 years.
• ESG Lab confirmed that the LUN and done simplicity of the XLR8r makes it quick and easy to deploy fast SSD capacity in a virtual desktop environment. A quick time to first access after power on of ten minutes was recorded.
• Applications ran without interruption as errors were injected, including the removal of a flash drive.

Issues to Consider

• Based on an audit of host-based performance counters and excellent XLR8r internal loopback performance results, ESG Lab believes that performance was host-limited during ESG Lab testing. In other words, testing with more servers would increase performance to, or above, the rated maximum of 250,000 random write IOPS (227,000 was recorded with the ESG Lab test bed documented in the Appendix).
• ESG Lab is confident that the price/performance advantages of the XLR8r will magnify over the next 12-24 months as WhipTail leverages the price declines and hardware advances of MLC flash drives, processors, and motherboards. For a proof point, ESG Lab audited the results of performance tests with the next generation of 6 Gbps MLC flash devices that are currently being qualified by WhipTail for a future product release. Internal loopback testing with 4K random IOPS running over only 12 MLC devices delivered 592,160 random 4K IOPS: more than double the performance of the currently shipping XLR8r.
• While ESG Lab has confirmed that the highly available design of an XLR8r SSD appliance can transparently survive hardware failures with zero downtime, WhipTail customers looking for higher levels of fault tolerance should consider the use of host level clustering (e.g., VMware Fault Tolerance, Microsoft Exchange Device Access Groups, Microsoft SQL Server Failover Clustering) for an added level of protection in the unlikely event of a failure of an XLR8r controller or logical data corruption at the application level.  WhipTail has advised ESG that a multi-controller design with enterprise-class fault tolerance and scalability is planned for a future release.

The Bigger Truth

Simplification tops the list of factors driving the adoption of VDI technology. 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 the respondents to an ESG survey indicated that reducing capital and operational expenses is driving an interest in VDI.

When ESG asked its survey respondents about current usage of and plans for VDI solutions, a significant 21% said their organization currently has a VDI initiative (in the form of an active production or test implementation), with an additional 8% saying they have plans to deploy the technology. Thirty-one percent of respondents said they had no current plans to deploy VDI, but would consider the technology. Since VDI technologies are relatively new, the fact that 21% of organizations surveyed already have some initiative underway is a significant development.

While the benefits of VDI are compelling and adoption has ramped up quickly over the past couple years, ESG research indicates that performance is a top concern. As a matter of fact, IT managers ranked performance (application response time) as their second largest challenge when it comes to implementing desktop virtualization.[5]

A modern processor can execute over a million instructions in the time it takes to perform a single random hard drive access. For IO-intensive virtual desktop application workloads that are mostly random in nature and tend to have more writes than reads, this difference in performance has caused a number of problems in IT organizations around the globe. What good is the ability to execute a million instructions per second if the processor is waiting for a disk access that takes hundreds of times longer? The traditional approach of adding more hard drives to solve the performance problem increases the cost, space, power, and cooling requirements of the storage solution.

A WhipTail XLR8r uses flash drives instead of hard drives to narrow the gap between processor and IO access performance. MLC flash drives are used to deliver cost effective price/performance that’s ideally suited for virtual desktop workloads. ESG Lab has confirmed that WhipTail Tech has turned an array of cost effective solid state MLC flash memory devices into a durable, highly available, high-performance storage solution that writes faster than it reads. Validation testing has confirmed that a single XLR8r appliance that takes only 2U of rack space and consumes less than 200 Watts of power can be used to support more than three thousand virtual desktop users with excellent response times, longevity of ten years or more, and LUN and done simplicity that makes it quick and easy to deploy.

WhipTail Tech sold its first MLC-based SSD solution in 2009. Since then, dozens of organizations have come to rely on the high speed and reliability of XLR8r appliances to eliminate IO performance bottlenecks for a number of business critical applications including databases, e-mail, and VDI. ESG believes that this is a key strength and advantage compared to the plethora of SSD startups that have emerged recently. The WhipTail team understands SSD application requirements and has years of experience supporting the latest MLC flash drives in production environments. If you’re planning a VDI deployment or you’ve run into an IO performance problem that can’t be met cost effectively with your existing disk array, ESG Lab recommends that you schedule a call with the SSD experts at WhipTail.

Appendix

[2] Configuration details are listed in the Appendix.

[3] 100% random, 80% write, 16 KB IO requests running on iSCSI connected servers with 64 outstanding IOs per device.

[4] Based on the assumption that a traditional mechanical disk drive supports between 100 and 250 IOs per second.

[5] ESG Research Report, Virtual Desktop Infrastructure Market Trends, February 2009.

Judging by the progress in this area, centralization appears extremely successful but alas, one especially difficult technology remains anchored to branch and remote offices–PCs. This is a real problem since PC management and security has always been a “one step up, two steps back” proposition. Remote PC security alone remains a bear. In a recent ESG Research report, IT professionals were asked to define their biggest information security issues around remote/branch office support. Managing remote PC security and configurations topped the list of challenges. Obviously, persistent PC security problems remain.

So what are large organizations doing to address this? You may be surprised to learn that a growing number of firms are ready to can the whole PC enchilada and replace physical remote PC configurations with desktop virtualization alternatives. In fact, 25% of organizations are already using desktop virtualization technologies to serve remote/branch workers, 22% plan to do so in the next 12 months, and 20% plan to do so in the next 24 months.

This makes a ton of sense–standard desktop images, centralized configuration and patch management, server-based storage of sensitive PC data, etc. I can see desktop virtualization gaining momentum in the future as Bring Your Own Device (BYOD) policies evolve further. What else does change mean? Here are a few thoughts:

1. Desktop virtualization represents a new use case for WAN optimization vendors. The primary job of WAN optimization controllers is accelerating Sharepoint, Exchange, and file access. Lots of vendors do this pretty well but desktop virtualization requires new protocol support and may open the market for new equipment or new vendors. Citrix is keenly aware of this and is already tightly coupling XenDesktop (and XenApp) with NetScaler.
2. Virtual desktops may not come from corporate HQ. My guess is that many firms will look at the transition from physical to virtual desktops and at least investigate SaaS provider options as an alternative to owning all of the servers, storage, networking equipment, etc. This may mean that branch offices become dual-homed with WAN connections to corporate data centers and direct Internet connections for cloud connectivity. This could be a profound change to typical branch office networks.
3. Virtualization will likely spread to tablets, smart phones, and remote workers. If virtualization can ease endpoint management and improve security, why stop at remote office Windows PCs? More and more endpoints will simply render graphics over the network rather than receive software updates and store sensitive data on local memory and disk.

You can read Jon’s other blog entries at Insecure About Security.

Further detail can be found in ESG’s Cloud Computing Trends Research Report.

VMware Horizon App Manager is best thought of as a broker of applications and initially a way for IT to broker SaaS applications through a common identity and authentication model. But the potential of Horizon App Manager is much more interesting. Brokering SaaS applications has some great benefits and may be all that is required for business that truly adopt a 100% SaaS model (there are few), but what about the other applications that enterprises consume? Desktop images? New applications? VMware Horizon App Manager can potentially mature and help here as well. I’d even go as far to think Microsoft RDS (terminal services), VDI images (VMware, Citrix, Microsoft, Quest, and others), applications (ThinApp, XenApp, App-V) and let’s not forget about the applications VMware themselves are beginning to stockpile (Zimbra, Mozy, Sliderocket … there are more coming).

So although VMware Horizon App Manger may be focused on SaaS brokering, identity, and authentication today, I suspect we will see it extend its breadth of capabilities to capture multiple application delivery models in order to truly deliver user centric computing. It will have to.

Read more of Mark’s blog entries at  Liquefying IT.