VMworld 2008 wrap-up – Part 1

This year’s VMware conference, VMworld 2008, broke all records with over 14,000 delegates and over 200 partners / competitors exhibiting.

As already written, understanding what really happened it’s impossible when looking at every single press announcement on its own.
There was a big need to collect all data, digest it, and try to figure out where the virtualization industry is going.

VMware is going away, on clouds.

As Micheal Keen, Mike Laverick, Massimo Re Ferrè and many other influencers have noted on their reports of the event, VMware pushed more than enough the cloud computing vision, clarifying that it wants a piece of the pie that Amazon has cooked and it’s eating all alone.

Of course, knowing that the company is working to drive the IT to the perfect world of the on-demand, ubiquitous computing it’s pleasing and exciting
A couple of years ago virtualization.info wrote that virtualization is just the first step of a long walk called grid computing.
We also published an article predicting that VMware would move on clouds as soon as Microsoft would step in the virtualization market with some concrete investments (and this because Microsoft wouldn’t be able to follow VMware on clouds for another ten years, giving the EMC subsidiary enough oxygen to stay ahead of the competition).

Nonetheless the message sounded very weak.

We need a vision. Made by human beings, the market ecosystem has to look forward and focus its energy on new technologies while selling/adopting the current ones.
In the past conferences, when VMware failed to tell its vision we complained and demanded for more. But there’s a limit: going too forward may confuse what is possible (and desirable) with what is just marketing hype-able.

In the Q&A session following his opening keynote, Maritz said that the company expect this future to concretize by a couple of years. From here it seems much farther away, like eight-ten years from now.

And looking at those 14,000 delegates we wonder how many of them were SMBs (let’s say 50% ?) and how many among those SMBs really care about embracing the cloud computing approach.

VMware is sure that SMBs will find such revolutionary architecture attractive because, at the first approach, it will offer them a cheap way to implement disaster recovery sites.
But we see a number of small companies still ignoring what capacity planning is, still struggling with P2V migrations, still figuring out if virtualization adoption really saves their money, etc.
More than that we see poor quality work done by many solutions providers that undermine the confidence in the technology and eventually drive the small firms back to the physical world.

Let’s skip the discussion about enterprises, where regulatory compliance and clouds can’t actually stand together in the same sentence (and hardly will do in the next two years).

In other words, if could computing is Virtualization 3.0 (as some over-zealous PR folks are already pitching), it seems that many companies are still at 0.9.

But let’s put aside for a minute any discussion about the feasibility of the vision and its go-to-mainstream timing.

This year’s VMworld was very special as the company:

  • has a brand new CEO
  • just lost both its co-founders (the CEO Diane Greene and the Chief Scientist Mendel Rosenblum)
  • is losing some key executives
  • is facing a much negative performance at Wall Street
  • is facing harsh competition (perceived or real it doesn’t matter) from Microsoft

In such impressive conjunction of events, one would expect the new CEO Paul Maritz coming on stage and telling to the company’s customers who he is, why he’s the right person to lead the company and that, at least, he has a solid plan for the next couple of quarters.

It’s early indeed: he took the new position just two months and a half ago, but not even a mention about the upcoming strategy?
Maritz repeated 35 times the word cloud (as somebody was picky enough to note) and couldn’t even spend a single sentence on how he plans to solve the problems that his company has today.

After his keynote VMW stock value had a significant drop. Is it legit to wonder why?


Luckily, the engine behind so many impalpable clouds is ESX, and the ESX 4.0 sneak preview that Steve Herrod, CTO, showed on stage was both concrete and thrilling.

With this upcoming release, VMware will demonstrate more than ever how virtualization is an amazing tool that can transform every segment of the IT.

If a company can afford the price (and this is the only point of the whole question, since ever), then VMware Fault Tolerance (formerly continuous availability), VMsafe, AppSpeed, Virtual Distributed Networking, and a number of additional features that stay unveiled for now, will introduce some serious innovation in its data center.

Some of these things may be still in early stage: as Rich Miller and Christopher Hoff promptly noted, the security ecosystem was surprisingly silent during VMworld around VMsafe, and Cisco couldn’t even announce the beta of its upcoming virtual switch for the Virtual Distributed Networking.
But even delivering in mid-2009, the VMware’s competition seems so far away right now.

Unfortunately this has a downside: the more VI 4 shines on stage, the more concerns it raises on the exhibit floor where 206 partners are telling stories about synergy and integration.

vCenter 4.0 (formerly VirtualCenter) is about to include a series of new modules that cover pretty much every need in the virtualization space, from chargeback to orchestration.
Every company that is trying to add value on top of the VMware management layer, may soon have a hard time to justify its existence.

Of course VMware reassures the partners promising flexible APIs and a friendly, cooperative go-to-mar
strategy, but the partners still remember the terrible experience with VDI in 2007, when the company first fed and promoted a healthy ecosystem and then killed it by acquiring one of the players (Propero).

On stage Steve Herrod basically previewed how VMware is about to invade all the spaces, and it’s hard to believe that a customer would prefer dealing with two vendors (which implies different licensing and support agreements) when it can have all the features he needs from the market incumbent.

So now the partners basically have two choices: stay loyal to VMware and focus their R&D investment on vCenter, in the only hope that they will develop some groundbreaking capability worthwhile of an acquisition, or re-tune their efforts for a more cross-platform strategy, where Microsoft (and the lack of features of Hyper-V and SCVMM) suddenly appears more attractive than ever.

Easily to guess, over the long term every vendor would look at Microsoft platform with interest. But nobody has unlimited resources to invest in development.
A vCenter that does everything just facilitates all those partnership deals with Microsoft that could be postponed by a year or two.

Relevant VMware press announcements:

In the second part of this wrap-up we’ll see where the rest of the virtualization industry is going while VMware is headed to the clouds.