Oracle has announced their own Xen based virtualization solution called Oracle VM. Finally Oracle is recognizing that their customers want and deploy virtualization solutions, but bringing "Yet Another Xen" (YAX) solution to the market seems to me to be a bit strange. How many do we have now? 5? 6? In addition, refusing to offer support for other virtualization products like VMware ESX server seems like a bad move.
What Oracle fails to understand is that a great number of their clients already run their products on other virtualization solutions and will be hard pressed to move their virtual machines to a "new" platform. They have already invested time and money in their infrastructure, and bringing another virtualization solution in to the datacenter is not something most clients will do in a heartbeat.
In the same way that Microsoft doesn't seem to understand that most virtualization customers want a OS-agnostic virtualization vendor, Oracle still fails to understand the basic virtualization needs of their clients, as they have for a long time with regards to their licensing policies.
From the Oracle VM FAQ:
Will Oracle support customers who are using Oracle products on other x86 server virtualization environments?
Oracle VM is the only x86-based server virtualization environment on which Oracle products are supported.
Bad move Oracle. Your clients won't like to be locked in like that, especially the ones that have already implemented virtualization in their infrastructure.
Until the hypervisor is an embedded technology, like the BIOS, customers will want freedom of choice when looking for virtualization partners. I'm pretty sure that in the future the hypervisor will be embedded in the hardware, and this discussion will be pretty moot, but for now it's not.
As mentioned before, Microsoft wants the hypervisor to be a part of the OS, Oracle wants the hypervisor to be a part of their product. I think both of them are dead wrong in their assessment of this, I think clients wants to separate their virtualization efforts from their other software providers.
Until now, I had this sneaky feeling that Oracle would acquire Citrix at some point, but this move by Oracle probably means that my predictions were wrong. I can see why Citrix bought Xensource and integrate Xen into their portfolio, it fits into their application/service delivery strategy. Incorporating the whole Citrix infrastructure into Oracle would also make some sense, at least in my eyes, but just doing a rebranded Xen version and then only offering support for their own version seems outright stupid.
One thing that this move highlights though, is the need for a unified virtualization management solution. There is definitely a market for a "one management solution to rule them all" strategy here, it will be interesting to see who comes up with something like that first.Update:
Chuck Hollis, from EMC, has posted "Oracle's Virtual Play" where he does a way better job than me outlining some of the problems with the Oracle VM release.
Broadly speaking, I think many parts of the IT industry has figured out what they're going to do with virtualization. They realize it's a big deal that means a lot to their customers, and changes all manner of things going forward.
Unfortunately, I don't think Oracle is one of them.
Yet.
And that's too bad.
I couldn't agree more.
0 Comments so far