What are people in the blogosphere saying these days about desktop virtualization? Actually, before we ask that, we should ask if there is really such a thing as the blogosphere anymore. In any case, there are people who write blogs and they have been writing about VDI and desktop virtualization, and here are three of the more interesting posts we’ve seen in the past couple of weeks.
There’s an piece by Dan Kusnetzky on ZDnet that discusses Intel’s approach to what it calls “Intelligent Desktop Virtualization” based on a conversation with Lisa Watts, Director of Business Client Solutions at Intel (Intel’s concept for intelligent desktop virtualization). The article notes that Intel’s concept of intelligent desktop virtualization is “using virtualization technology to create an environment that has the following characteristics:”
Companies that offer products that are examples of Intelligent Desktop Virtualization: RES Software, Virtual Computer, Wanova, MokaFive and Scense.
Here’s a real interesting post from Spankmeister at Spiceworks (Virtualization: What’s next for the enterprise? VDI, that’s what). This one was just posted yesterday and already has 50 replies. Spankmeister wrote about his own experience using VDI. He started by researching a couple of vendors that support VDI; made sure it worked with his existing infrastructure; got a couple of test devices, and tested on both LAN and remote locations. Here are some of his conclusions, based on what’s he’s done so far:
If you’re thinking about deploying VDI, or if you’re in the process of evaluating, it will be well worth taking a look at this post and also looking at the various comments.
Ken Hess, also of ZDnet, expresses why he’s had a change of heart about VDI, from being strongly against to being a fan of certain types of desktop infrastructure implementations (Is VDI Really an Option?). Why the change of heart, he asks himself: Timing and technology, he answers. “VDI really was not an option before,” he says, “But it is now,” nothing that it’s time to look at VDI again in a serious and more practical way. Hess suggests that you convert a few dozen desktops as a test. “The test will provide feedback about how well your employees work with virtual desktops in each area of your business,” he suggests. “Some departments will prove easy to convert, others will be difficult due to user issues, a few will have correctible technical difficulties and a very small percentage will not be able to make the switch.”
Hess has an interesting take on the future, particularly for someone who was not an advocate of VDI for a while. “Is it the future of desktop computing,” he asks himself again. His reply: “For 90 percent of us, yes. For the other 10 percent, it’s a little further into the future.”
What do you think? Is VDI, or some form of Intelligent Desktop Virtualization, as Intel describes it, the future of client computing? If you have an opinion, feel free to comment on the site. Even if you don’t have an opinion you can comment on the site anyway.
What’s going on with virtual containers as a desktop virtualization strategy?
Intel had been – and presumably still is – a major proponent of virtual containers as an alternative to hypervisor-based VDI solutions. I was looking through some older marketing material from Intel and they make a strong case, but there seems to be a lot less momentum in virtual containers than in VDI.
With virtual containers you basically have the operating system and applications created and managed centrally by IT and then streamed to the client, where there is a client-based virtual machine manager (VMM). One of the major advantages of this approach is that it doesn’t require the same type of infrastructure investment as VDI because much of the workload takes place on the client side. Another advantage is mobility, since users can cache their virtual containers for off-network use, then hook back up to the network.
It is my understanding that virtual containers typically mean the use of one operating system, which is a disadvantage versus VDI, but I am not 100 percent sure if this issue has been addressed or is being addressed. Another potential disadvantage is that interactions between applications may be limited.
So what’s the story with virtual containers and what impact will they have in the world of desktop virtualization? I have a feeling someone from Intel may be reading this blog, so perhaps we can generate our first comment(s). Yes?]]>
We were thinking about the impact of different approaches to desktop virtualization in addition to VDI and we came across this interview from last week with Stuart Dommett, Business Development Manager at Intel, conducted by Tim Phillips of The Register in the U.K. Here are some interesting points:
Make sure you build the right infrastructure: One of the challenges of VDI and desktop virtualization in general is in understanding the impact and the costs on the rest of your infrastructure. No doubt you will need to bolster your networks and storage capacity, but how much? Dommett says you should go back and profile both your applications and your users in the context of: What tools do they need, what is their job and how are they going to work. “If a mistake is made, these are costly mistakes,” he warns. Not having the right infrastructure not only impacts you financially, but also in user costs and productivity.
VDI is one solution, but not the only one: IT decision-makers are looking at using a wide range of technologies, such as terminal servers and “baking it in with some VDI because of some specific needs or specific segments of their installed base. Desktop virtualization has grown further as understanding and tools have matured and developed. Application virtualization, or the ability to deliver applications into various environments, is opening up VDI, terminal services and is really pushing applications to endpoint devices.”
Has VDI been oversold: “VDI has its place within an environment, but you have to question why you’re doing it,” Dommett suggests. “With VDI you’re processing everything on the server. I’ve talked to some of my server colleagues and the question is why would you run a PC on the server? Is that the best use of your server performance? They don’t have spare capacity, so why would you go and put more and more into the data center? Why is that the best solution when what you can do is look at some of the techniques, centralizing your application, centralizing your image build, making sure your data is protected or kept within the environment through other types of client or desktop virtualization. We don’t have a true answer yet on the success of VDI. A lot of customers are going down that path and time will tell if it is the right investment or the mature model that will stay in the market for the next eight to ten years.”
What’s next: One of the big trends to watch is in the adoption of consumer products and how those technologies will end up impacting the whole area of client virtualization. However you call it, Dommett sees desktop virtualization as one of the two to three biggest trends for IT moving forward.
What do you think? Keeps those cards, letters, emails and comments rolling in.]]>
So today we start our VDI Trender blog and we feel compelled at the outset to state that VDI Trender is a bit of a misnomer in that VDI is one of the desktop virtualization solutions we will be talking about here, but not necessarily the only one. As we get talk to more and more people about VDI, we believe that VDI is often used as a catchall for a variety of different types of solutions. It is also the one that receives the most attention because it is the one that the big-name companies – i.e., VMware and Microsoft, to name the most prominent– tend to use most often in their marketing efforts. It is also the one that the pundits such as Gartner and Forrester have identified as “the next big thing” in client computing. But VDI, or Virtual Desktop Infrastructure to be complete, is but one approach to desktop virtualization and we shall attempt to talk about all of them here.
We thought about this as the starting point for our blog because of a couple of things: (1) In our most recent research into the topic we came across a survey by CDW that we thought should get more attention because of some of the numbers (quite compelling) and (2) because of an interview we listened to over the weekend with an Intel executive in the UK talking about some of these very same points. We’ll focus on the CDW research today and the Intel interview in our next post on Wednesday.
In September 2010 CDW surveyed 200 managers at medium and large businesses to understand what they describe as client virtualization – note how they avoid using the specific term VDI. We thought the numbers were pretty impressing – 90 percent of those responding to the survey said they are either implementing or considering implementing at least one form of client virtualization. The most critical driving factor was reducing costs, followed by easier software distribution, increased IT productivity and improved IT support. Of those that have already implemented some form of client virtualization, the average savings are slightly more than 20 percent of their total IT budget.
When CDW asked about client virtualization, they attempted to divide the types of solutions into three main categories, which gets back to my early point about what exactly is VDI and how does it fit in as a defined term by IT professionals. This is the language used by CDW for its client virtualization categories:
Presentation Virtualization: Enables users, usually remote workers, to use the organization’s network and, if authorized, access all applications that reside within the corporate data center through a Web-based portal.
Application Virtualization: Describes applications that are packaged into single executables that run completely isolated from each other and the operating systems for conflict-free execution on end-point devices.
Desktop Virtualization: Brings the desktop operating systems and applications into the data center, so that all of the processing power is hosted in the data center. As long as users can access the Internet, they have access to the business applications.
Note that even in the “Desktop Virtualization” definition, the language used is not VDI or Virtual Desktop Infrastructure.
We think a couple of things:
1. This is way more confusing than it needs to be. Not sure if it’s because of the hype surrounding VDI as a term or the lack of clarity in various solutions, but markets tend to jell around topics and solutions people can actually articulate and understand.
2. Who are these 90 percent of IT managers are what are they actually doing with client virtualization? That is quite a large number and much higher than you would expect based on actual deployments and/or sales – particularly considering that the survey was fielded about eight months ago. My sense: Yes, a lot of people are “considering” client virtualization because they are supposed to be considering client virtualization.
What do you think? Is the language in this market more confusing than it needs to be? Are 90 percent of companies really ready to do something about client virtualization? And how about VDI – what does it really mean to you? Please comment and share your thoughts. And, if you want to check out the CDW Survey: CDW Client Virtualization Straw Poll.