Now that we’ve had a few weeks to recover, I figured I could finally take the time to answer the question that everyone’s been asking me: “What did you think of VMworld?” That question really needs to be answered in two parts: (1) What do I think of VMware after the show, and (2) what did I learn about the other vendors in the exhibit hall? In today's article I'll answer the VMware question, and tomorrow I'll summarize the cool stuff I learned about the other vendors.
So with regards to VMware, I'll come right out and say it: Everyone knows that I was nervous about VMware's desktop strategy before VMworld, and now that VMworld is over I'm just as nervous about it as I was before. I just don't think desktop is a real focus for that company.
I can't say that I entirely blame them. It's understandable that VMware would want to diversify after their first big success in the server space, and I think that's what led them to the desktop in the 2006-2007 timeframe (Propero, Thinstall). But now that they got a taste of cloud it's almost like, "Desktops? Oh right, we do that too."
I'm not saying that VMware shouldn't evolve their focus or that they shouldn't go after the cloud. (And frankly the whole SpringSource thing plus their virtual infrastructure to go after Microsoft as an app platform seems pretty brilliant to me.) I'm just saying that even after VMworld, I don't believe that VMware has demonstrated that they have the capability to be a big player in the desktop space too.
The irony of this is that I really, really want VMware to be successful with desktops. Competition from VMware has caused Citrix to put more innovation into their desktop delivery products in 18 months than they have in the past ten years combined. If VMware rolls over on the desktop, I hope that the Symantec or Quest can step up to apply pressure to Citrix or else we're going to have another decade of innovation ice age.
But back to the main topic: VMware's lack of desktop vision.
At last year's VMworld, we got all these great indications of what VMware was capable of in the desktop space. (Here’s my article from then where I could hardly contain my excitement about six desktop announcements.) And how did VMware follow up at this year's VMworld?
- We got demos of the PC-over-IP and the client hypervisor. (This has a lot of promise and looks to be really great.)
- We learned that VMware will OEM RTO Software’s Virtual Profiles technology for some future release of View. (Not View 4 though, from what we can tell.)
So those were two great things. (Tactical, but still great.) Unfortunately that's kind of where the good news from VMworld stopped:
- VMware demoed their client hypervisor known as CVP. (yay!) But we learned that it won’t be available until 2H 2010! Last year they said their client hypervisor was going to ship by the end of 2009, and now it's another 8+ months away!?!
- The View futures session was just about View 4. VMware still hasn't announced a release date, although the rumor is it's still coming at some point this year. View 4 will have PC-over-IP, but little else new. (Oh, it will have vSphere 4 support.)
The big glaring hole was some frank talk about what VMware's actual desktop strategy is. When I wrote that rant-ish article a few months ago, I accused VMware of just riding their server coattails into the desktop space without a real clue on what desktops are and how they'll compete with Citrix or Microsoft. And now that VMworld is over, I still don't know how VMware plans to compete?
So that said, let me be more specific. Instead of whining about a fairly esoteric question of "what's your strategy?," let me ask VMware to address thirteen specific questions / concerns that I (and others?) have in the desktop space. (And thanks to BrianMadden.com member "appdetective" who's responsible for four or five of these questions based on various comments he or she posted to previous blog posts.)
- Paul Maritz said that VDI is a slow ramp up before it represents revenue for VMware. In this resource- and budget-constrained economy, what's their commitment to this space? Why should I believe anything they say after comments like that from their CEO?
- Is VMware going to support a non-ESX solution for VDI? If not, why should I be locked into their hypervisor solution? (Hello Freedom anyone?)
- When Microsoft releases Calista, they're likely to do so only for Hyper-V. What does that mean to all the people who are deploying View with ESX only?
- Why does VMware think they can win a battle of the desktop against Microsoft when Microsoft owns the operating system? Isn't that reason enough that smart customers should sprint towards Microsoft+Citrix or Microsoft+Quest?
- Who's actually deploying View at scale? It seems that it's fine for small ESX-only shops that are LAN-based and don't require any medium to advanced features. Who's adopting View for Cloud DAAS? What is the size of their largest customer deployment?
- vSphere 4 was released last June. But View 3 is not supported, and we still don’t have an official release date for View 4. How is this possible if the desktop is really a priority? How are customers supposed to sit through six months of marketing about how great ESX 4 is, but then not be able to leverage it for their View solutions?
- Doesn’t VMware have a vested interest in moving the world off of Windows applications? (SpringSource) How will we expect VMware to create the best system for delivering Windows desktops while also creating a system to replace them?
- Does VMware think that ThinApp can handle all apps today? What is the official recommendation for people who have apps that aren’t ThinApp compatible? Should they all be installed natively? Should they be virtualized with another product? Should View only be used where we have 100% ThinApp compat?
- What’s VMware’s recommended technique for deploying ThinApp apps into VMs? Let’s say I really want to use View for everyone. How are apps delivered and managed?
- Does VMware believe that there is any use for server-hosted apps which could be delivered seamlessly into VMs? If so, what’s their recommended product for that? Citrix XenApp? Quest vWorkspace? Microsoft RemoteApp?
- What are VMware’s thoughts about the role the client hypervisor will play? Is it truly an “offline VDI” solution?
- Why is CVP delayed? (And even when it comes out in 2H10, it will still only be a single VM? So it's super late and won't even be fully featured when it does come out. (And at that time, competing products will have been on the market for a full year that offer multi-VM support.)
- VMware's new desktop CTO is Scott Davis, one the founders of Virtual Iron and most recently VMware's chief datacenter architect. In other words, a server guy is VMware's desktop CTO, furthering the image that VMware thinks desktops are just like servers, and in fact they really don't understand them. Scott's own LinkedIn profile says, "A recognized expert in virtualization, clustering, operating systems, file systems and storage." (i.e. no desktops.) So what are Scott's desktop creds? How do we know he's the right guy to lead the desktop strategy? Why did VMware hire a server guy from within instead of a recognized desktop expert?
So now what? I guess ideally I'd love to get these questions and concerns addressed. I don't know how that will happen, though. Maybe we can get Scott and Jocelyn Goldfein (VMware's general manager of the desktop business unit) on Brian Madden TV to talk about this stuff? (Actually, screw BMTV. Maybe we can just talk openly for a few hours and get these questions answered.)
What do you think? Am I off base here? Does anyone reading this know any of the answers to any of these questions? Are there any more questions that you'd like VMware to answer about their desktop strategy?