Notes from VMworld: ignoring phase 0

So the first two days of VMworld so far have exceeded my expectations. It wasn’t so much the sessions (though they were pretty good). It wasn’t the partner super session. The super sessions basically verbalized what is on everybody’s mind right now. And that’s the thing that makes it exciting: Its the vibe in the air. Everyone knows that there’s a big change happening in the industry. And with this big change, we all sense opportunities.
The cool thing about these opportunities is that the field is wide open. Certainly VMware has a giant leap on everyone as solidified by the magic quadrant they’re hyping on their home page right now. But the way VMware is set up to let partners also develop solutions based upon them gives them sizable advantage. However, there are other great things afoot as exibited by the vendors in the tech floor down below where I write this.

So here are the coolest things so far:

1. The labs: They took a risk, put the lab in the cloud. I remember trying to do something like this at IBM and thinking it opened up a whole new realm of possibilities. Now the labs don’t need to be given strictly at the conference. In fact there’s no reason that VMware won’t offer these labs outside of VMworld. A fantastic investment. They’ve had some snags: slow networks, machines not provisioning, etc. However, it is usable and they announced that 3800 labs were deployed.

2. Meeting people that we’ve been reading from and just hearing ideas from people that I’m sitting next to. This has been great and the ideas are just flowing. There are lots of brilliant people here that freely share ideas. I really dig this.

3. The ability to sign up and talk to technical experts: a 1 on 1 session for 15 min to get questions answered. This has been great.

4. Seeing all my clients, partners, and long time friends. This is really what its all about: Meeting the people, getting contacts, leads, and laughing about the 24 hour sleepless nights and adventures we shared X years ago.

But there’s also a lot of fud and marketing and statements made that I don’t agree with.

One of the statements I disagree with is what they said in the partner super session and one that has been repeated many times. They say:

“virtualization is stage one of any cloud”.

I strongly disagree. Coming from an HPC background, I am very adamant about the statement: virtualization doesn’t equal cloud. And for that manner, you do not even need virtual machines to have a cloud. Virtualization comes after you’ve got a handle on your data center. This includes switches, physical machines, etc. So if virtualization is stage 1, then stage 0 is getting control of hardware.

Stage 0 means a lights out data center where its dark because people only go in there once in a while to replace failed components. Stage 0 also requires the ability to re-purpose hardware on demand. This is something we’ve been doing in xCAT for many years. Stage 0 means we can power physical machines off / on. We can deploy hypervisors or native OSes without hypervisors to physical machines over the network. This requires a centralized deployment engine. And all of this is the bottom layer that we at Sumavi and the open source xCAT community have been working on for many years.

This area of functionality can not be trivialized and VMware gets it. However, there is no product in its portfolio to hype other than some beta works so the problem is largely ignored and religated to “Look to your hardware vendor to provide this solution”. However, this ignores multivendor sites, which is pretty much everyone.

All of this has given me the feeling that what we are doing at Sumavi is becoming more and more important. Our partners and our customers have stressed the need for it. You can be sure to see more products in this space as time goes on. And you’ll certainly hear more of it at VMworld 2011.

Comments are closed.