After completing my recent annual vacation in Crete last week and fortunately beating the dreaded "Ash Cloud," I was quickly back in the saddle. You know what it's like after some time away from work -- you have a mountain of email to work through, and you start to wonder how you will ever get on top of it all.
Fortunately, it was my pleasure to spend some time last week with VMware at their U.K. head office in Frimley, Surrey. As you might (or might not) know, I'm working on a new book about VMware View 4.5, and I spent two days with one of VMware's top SEs, thrashing through some questions about the soon to be released product. I found the time spent there extremely valuable; I went with a whole list of questions based on my use of the beta product over the last two to three months. Every question was answered, and you can't fault that!
Now, of course I can't go into any technical details about VMware View 4.5, as I'm constrained by a beta program NDA, but I want to comment on some of the speculation surrounding VMware View 4.5 that surfaced whilst I was lolling on the beach.
VMware View 4.5 delayed?
Firstly, there's been quite a lot of speculation that VMware View 4.5 is delayed. This was reported firstly on Brian Madden's website, and picked up by others. I'm not sure what Brian's sources are, but it seems clear that his post was prompted by having a briefing cancelled.
Additionally, there was a report by TechTarget's very own Bridget Botelho that View would be "pushed off its release date until sometime later in 2010."
Personally, I think this says more about how the bigger companies like VMware get, the harder it is to get their message across to the public clearly, without disclosing commercially sensitive information.
According to my sources, VMware View 4.5 is not delayed and will reach GA to schedule. It's true that an announcement was delayed, but the product itself should ship as planned. I was not told the exact date, but it seems clear that sometime in Q3 is likely. I think an unfortunate side effect of these rumors and speculation is that it bleeds into other aspects of the VMware View 4.5 release.
Secondly, whilst I was away, Citrix debuted their "client hypervisor" called XenClient. In case you don't know, it allows you effectively to install a hypervisor to a PC. From this PC, multiple virtual machines can be started. This is an important release for Citrix in that it shows commitment to R&D. Also, since they don't have an equivalent to VMware Workstation or Microsoft Virtual PC, it's important to show to their customers and themarket that they are as serious about client-side virtualisation as they are about servers.
The impression I get from both VMware and Microsoft is that the development of a client hypervisor is less a priority for them because they have offerings in this area already which are much more mature, and that the demand from customers is not as great as some commentators might like to imagine. It's perhaps salutary to remind everyone that the list of compatible devices for the Citrix XenClient is quite narrow (it requires a Intel based processor enabled for Intel VT-x Technology and VT-d Technology), and that it itself is a technology preview -- the product won't itself GA until later this year -- at the moment, it remains a trial/evaluation "express" product.
Unfortunately, in our industry there does persist an unhealthy obsession with GA dates. If they slip or change, many commentators are far too quick to cry foul because it makes good copy. The reality is that many customers won't even deploy a new technology such as VMware vSphere4 or Citrix XenClient until six to 12 months after the GA date. In fact, by the time these customers are ready to roll-out, there may already be a new version on the horizon. Such is the dizzying rate of releases and re-releases we receive from software companies.
This doesn't mean that VMware doesn't have a client hypervisor of its own, but that's it taken a more conservative approach in its development. I have a feeling that even if VMware could release their own client hypervisor today, they wouldn't. Primarily because there's little point in releasing a product that would have such a narrow HCL, for which sales would be miniscule, and which right now is not at the top of the agenda of most customers.
Editor's note: This is part one of Mike's column on recent data centre speculation. Read part two here.
ABOUT THE AUTHOR: Mike Laverick is a professional instructor with 15 years experience in technologies such as Novell, Windows and Citrix, and he has been involved with the VMware community since 2003. Laverick is a VMware forum moderator and member of the London VMware User Group Steering Committee. In addition to teaching, Laverick is the owner and author of the virtualisation website and blog RTFM Education, where he publishes free guides and utilities aimed at VMware ESX/VirtualCenter users. In 2009, Laverick received the VMware vExpert award and helped found the Irish and Scottish user groups. Laverick has had books published on VMware Virtual Infrastructure 3, VMware vSphere4 and VMware Site Recovery Manager.