I’ve recently been thinking a lot about application-specific workloads and architectures (Optimize Scalalable Workload-Specific Infrastructure for Customer Experiences), and it got me to thinking about the extremes of the server spectrum – the very small and the very large as they apply to x86 servers. The range, and the variation in intended workloads is pretty spectacular as we diverge from the mean, which for the enterprise means a 2-socket Xeon server, usually in 1U or 2U form factors.
At the bottom, we find really tiny embedded servers, some with very non-traditional packaging. My favorite is probably the technology from Arnouse digital technology, a small boutique that produces computers primarily for military and industrial ruggedized environments.
Slightly bigger than a credit card, their BioDigital server is a rugged embedded server with up to 8 GB of RAM and 128 GB SSD and a very low power footprint. Based on an Atom-class CPU, thus is clearly not the choice for most workloads, but it is an exemplar of what happens when the workload is in a hostile environment and the computer maybe needs to be part of a man-carried or vehicle-mounted portable tactical or field system. While its creators are testing the waters for acceptance as a compute cluster with up to 4000 of them mounted in a standard rack, it’s likely that these will remain a niche product for applications requiring the intersection of small size, extreme ruggedness and complete x86 compatibility, which includes a wide range of applications from military to portable desktop modules.
Bill Gates said "People everywhere love Windows.” Whether or not you agree, the fact that Microsoft Windows remains the de facto standard for business productivity after nearly 3 decades, suggests that many still do. But as the sales figures of Microsoft’s competitors suggest, people everywhere love lots of other things too. And one of the reasons they love them so much is that they like to get things done, and sometimes that means getting away from the office to a quiet place, or using a technology that isn’t constrained by corporate policies and controls, so they can be freer to experiment, grow their skills and develop their ideas uninhibited.
Technology managers I speak with are aware of this, but they’re justifiably paranoid about security, costs, and complexity. So the result of these conflicting forces coming together is inspiring rapid innovation in a mosaic of technologies that Forrester collectively calls digital workspace delivery systems. It involves many vendors, including Microsoft, Citrix, VMware, Dell, nComputing, Amazon Web Services, Fujitsu, AppSense, Moka5, and more. The goal of our work is to help companies develop their capabilities for delivering satisfying Microsoft Windows desktop and application experiences to a wide range of users, devices, and locations.
When I was maybe 2 years old, my mother lost track of me in a Toys-R-Us store. After a dozen stressful minutes, she finally found me - holding a Fisher-Price airplane. And so began my love affair with airplanes and aviation. So as I looked through the break-out schedule while attending NVIDIA’s GPU conference two weeks ago in San Jose, California, Gulfstream Aero’s session on transforming manufacturing and field service with desktop virtualization caught my eye. It didn’t disappoint.
There are 2 reasons why I liked this session so much and why I think it’s worth sharing with you:
It’s a nice example of technology that makes the work easier for employees, and helps them improve the customer experience directly.
It’s also an example of how a technology that’s not necessarily a money saver (in this case, VDI) shines when it enables workers do something that would be difficult or impossible any other way.
This week, IBM announced its new line of x86 servers, and included among the usual incremental product improvements is a performance game-changer called eXFlash. eXFlash is the first commercially available implantation of the MCS architecture announced last year by Diablo Technologies. The MCS architecture, and IBM’s eXFlash offering in particular, allows flash memory to be embedded on the system as close to the CPU as main memory, with latencies substantially lower than any other available flash options, offering better performance at a lower solution cost than other embedded flash solutions. Key aspects of the announcement include:
■ Flash DIMMs offer scalable high performance. Write latency (a critical metric) for IBM eXFlash will be in the 5 to 10 microsecond range, whereas best-of-breed competing mezzanine card and PCIe flash can only offer 15 to 20 microseconds (and external flash storage is slower still). Additionally, since the DIMMs are directly attached to the memory controller, flash I/O does not compete with other I/O on the system I/O hub and PCIe subsystem, improving overall system performance for heavily-loaded systems. Additional benefits include linear performance scalability as the number of DIMMs increase and optional built-in hardware mirroring of DIMM pairs.
■ eXFlash DIMMs are compatible with current software. Part of the magic of MCS flash is that it appears to the OS as a standard block-mode device, so all existing block-mode software will work, including applications, caching and tiering or general storage management software. For IBM users, compatibility with IBM’s storage management and FlashCache Storage Accelerator solutions is guaranteed. Other vendors will face zero to low effort in qualifying their solutions.
This morning Citrix announced the acquisition application mobilization vendor Framehawk for an undisclosed sum as the battle for high performance for corporate Windows apps on mobile devices rages on. Here’s my take:
It's a good acquisition for Citrix and in turn for I&O pros for 3 reasons:
Some of Framehawk's technology will be additive to Citrix's enterprise portfolio. Specifically, Framehawk's framebuffering protocol - called Lightweight Framebuffer Protocol, or LFP - is designed for mobile carrier networks like 4G/LTE where there is often highly variable latency, loss, and jitter. Citrix will add it to their arsenal alongside HDX to improve the end user experience of server-hosted Windows applications on mobile devices for XenDeskop App Edition and XenDesktop.
It will be a boon for DaaS providers' customer experience. Citrix is in the business of building a Desktops-as-a-Service (DaaS) platforms for service providers. One of the barriers to the success of DaaS in the enterprise, and a potential source of value for service providers, is the user experience on mobile devices over mobile networks. Another player to watch the remote desktop/app protocol space for mobile networks is RapidScale.
It's a competitive take-out play as well. Delivering Windows apps from the datacenter to both corporate and employee-owned desktops, laptops and mobile devices is what Citrix does - it's their place in the technology universe. Framehawk's technology approach, while expensive, has some advantages. Citrix was probably starting to see them in more deals as competition.
In Barcelona this week, VMware announced that it is acquiring Desktops-as-a-Service provider, Desktone. This is a market I've been watching for several years, and I think this is good news for both Desktone and VMware customers. On one hand it provides an alternative for VMware prospects who are unsure whether they want to make the investment in ramping up an in-house VDI initiative, and it provides a scale-out option for existing VMware View customers who may be loathe to make additional capital investments to expand their capacity. With Citrix also developing their own homegrown DaaS infrastructure offering for service providers, this move further legitimizes the DaaS market.
Forrester has been tracking the rise in interest in DaaS specifically in our Forrsights surveys of IT decision makers for the past 2 years, which gives us a unique view into the market. In Figure 1 below, we can see the rise in IT decision-maker interest in DaaS relative to on-premise hosted virtual desktops, and see that year-over-year growth of DaaS interest is strong. The market accelerated in part because Infrastructure-as-a-Service providers see it as a way to monetize their existing infrastructure investments.
Winning teams delight in the 'doing', not the 'winning'
When I was growing up in Pittsburgh, Pennsylvania, Chuck Noll was the head coach of the Pittsburgh Steelers, and his leadership was my first glimpse into what separates teams that win from teams that don't. My favorite quote from Noll exemplifies his view both simply and eloquently: "The thrill isn't in the winning, it's in the doing." What Noll taught his players, including Terry Bradshaw, 'Mean Joe' Green, and Franco Harris, is how to delight in the joy of doing, over the joy of winning. Why? Because the 'winning' won't come unless your passion comes from the 'doing'!
This is why I'm so excited about the explosive trend of BYOD and consumerization of IT on so many levels, from cloud computing to tablets. Workers are putting the joy back into the 'doing' parts of their jobs by exploring different ways of working, and using technologies that often exceed what even the best IT organizations can generally provide. It's also why I think we'll see a rebound of Hosted Virtual Desktops (also known as VDI) through 2013 and beyond. But first, let's look at the data:
IT interest in VDI appears to be on a downward trend
Forrester's annual survey of IT decision-makers revealed a drop in interest and plans for VDI initiatives from 2011 to 2012:
But a shift in the IT drivers for VDI suggests it could actually be an inflection point
End User Computing is at the Root of the VMware Family Tree
Examine the roots of the VMware family tree, and End User Computing is the longest root of 'em all. It's where it all began, back in 1999 with a cool little product that let me run Windows on top of Linux. It was like magic for software customer demos of complex enterprise apps. I could royally screw up a demo environment an hour before a demo for a $15M deal by adding just one field to the screen that the customer demanded to see, but instead of soiling my underwear in a panic, I could go back to my most recently saved state of less than an hour before. Brilliant! It was a tool for me to be more effective in my job. Hold that thought.
So with this heritage in mind and a general respect for VMware's products honed over the past 15 years of growth and change, and fantastic tools for I&O professionals to manage virtualized environments with, I was delighted to see End User Computing be the focus of general session demos and breakout sessions. I was looking forward to learning more about Wanova Mirage to see if it could help on the employee freedom and personal innovation front. Those of you following this space know what I think of what I like to call Soviet Bloc Virtual Desktop Infrastructures.
Virtuosity as the Root of Innovation and the Dangers of Hosted VDI
At Forrester, each of us as analysts keep in regular contact with our clients and the industry through a process known as Inquiry. For workforce computing, this includes Benjamin Gray, Christian Kane, Michele Pelino, Onica King, and Chris Voce. Any Forrester client with Inquiry access can arrange for 1:1 time with an analyst to ask questions and seek advice, or simply ask for a response by e-mail. Most analysts also take advantage of the opportunity to ask a few well-considered questions of our own. Taken together with data, briefings from vendors, ongoing research and client advisory, the inquiry process helps us keep our eyes and ears focused on what matters to I&O professionals, and provides critical insights into their pain and needs. In this blog, I'll share my unvarnished responses to a client inquiry I received just last week:
1. What do you see as the most important trends in End User Computing for the next 3-4 years?
2. What will be the role of each type of device in an organization such as ours (financial services)?
3. What's the best way to find out what our employees need? What do other firms offer different types of workers?
4. Do you have any economic numbers about those devices (i.e. TCO per year)?
5. Do you have any data or examples from other firms like ours?
Demand for mobility is rising dramatically, but IT support is not keeping up. Over the next 12-18 months, we expect a majority of Asia Pacific (AP) organizations to begin to feel the pain of poor mobility strategies. Now is the time to define and manage mobility as part of a broader end-user computing strategy – this must include desktop virtualization initiatives, including (but not limited to) virtual desktop infrastructure (VDI). But while server virtualization is now accepted as a fundamental design principle and part of any data center implementation or refresh, that doesn’t mean desktop virtualization will follow suit. Long touted as a means to simplify desktop provisioning and management – and hence improve the efficiency and effectiveness of an organizations’ end-user computing strategy – over the past decade desktop virtualization has been driven primarily by CIO’s desire to lower hardware costs – by delaying or skipping PC refresh cycles – simplify application provisioning, and increase compliance and control of desktop infrastructure in areas like data security and patch management. Desktop virtualization doesn’t adequately address all end-user computing requirements since it’s essentially focused on eliminating the client device from the equation. This is particularly true for VDI. Thin (e.g. ‘dumb’) clients won’t work in a world where a growing percentage of users – not just information workers – are mobile and expect access to key resources but also expect those resources to be optimized for the particular device they’re using. With the explosion in device usage and changes in end-user expectations, IT is being forced to expand its focus around end-user computing from ‘control’ to ‘engagement’. Desktop virtualization will remain a key component of many organizatons’ end-user computing strategies, but its role will remain