Cameron Haight

Like several of my colleagues, I am at VMworld today.  The keynote was provided by Paul Maritz, the new VMware President and CEO.  The delivery of his presentation was understated, but the ramifications could be very significant to the IT industry.  I use the word “could” because it’s an ambitious agenda, but as Maritz said to many of us, VMware had to decide if they were going to become a platform or a feature … and they chose the former.

 

So, what was announced?  Three major initiatives: Virtual Datacenter OS (VDC-OS), vCloud and vClient (I am going to leave discussion of vClient to my client computing colleagues at Gartner).  VDC-OS, is the follow-on to VMware Infrastructure and is similar in concept to the meta-OS that my colleague, Tom Bittman (http://blogs.gartner.com/thomas_bittman/), has written about in past research (way to be forward looking, Tom).  It was defined as an elastic, self-managing, self-healing “substrate.”  One of the key points brought out was that VMware was thinking about future application loads and usage scenarios … and wanted to provide a more robust environment that would remove some of the uncertainty of today’s virtual infrastructures regarding application QoS. 

 

You will also see much more usage of the letter “v” in front of the names associated with this and the other key initiatives.  As part of this theme were announcements of new Infrastructure vServices and Application vServices.  The former focuses on things such as new storage capabilities (thin provisioning, linked clones, etc.), network switching (including a new supporting Cisco product) and direct VM to hardware access (I am glossing over these, but there are some interesting management implications associated with these new capabilities which will be discussed in the future).  Application vServices includes new functionality such as VMware Fault Tolerance and VMware Data Recovery to deal with unplanned outages.  One of the more interesting components of Application vServices is vApp, which is leveraging the Open Virtualization Format (OVF) to provide a means to better describe the operational and service level requirements of a multi-tier application (note: if you are a Gartner client, you can read more about OVF in a note that I co-wrote here: http://www.gartner.com/DisplayDocument?ref=g_search&id=619109). 

 

There’s also Management vServices.  VirtualCenter becomes vCenter – and will be more open (i.e., easy to integrate with third party applications via the VMware Ready Program for Management Solutions) and more scalable (and highly available).  Several new management offerings that will be bowing in this area include chargeback (vCenter Chargeback), capacity planning (vCenter CapacityIQ which leverages some of the design concepts of VMware Capacity Planner) and policy-based configuration management (vCenter ConfigCheck).  The recent purchase of B-hive is becoming vCenter AppSpeed and will be VMware’s eyes into application response times, etc.  Dunes, which was purchased last year and was launched earlier this year in the form of VMware Lifecycle Manager, will provide another spin-off into vCenter Orchestrator which will deliver a more generalized workflow or run book automation system. 

 

Finally, there’s vCloud.  vCloud seems to be something that’s still very much in the design phase, but was generally described as a series of technologies to enable the federation between enterprise and off-premise clouds.  It will leverage the VDC-OS, vApp as well as a series of APIs that will enable a common means by which to invoke off-premise cloud services.  While no firm dates were available, it appears that many of these products and technologies will be rolling out over the next several years (and perhaps beyond). 

 

So what’s the verdict?  Of course too soon to tell.  Is VMware overreaching with the VDC-OS?  Candidly, as I was listening today I had thoughts of other somewhat similar  “autonomic” initiatives, i.e., IBM’s On Demand, Microsoft’s DSI, etc.   To date, these have not turned out so well.  However, from an architectural (and street cred) perspective, one may be able to argue that VMware is perhaps in a better position to deliver on this vision than some of the others.  Still, it’s probably best to temper the enthusiasm until we see more proof points.  Same as well for the new vCenter add-on management offerings.  We need to give VMware some more time to wrap them up and then re-visit them later in a more detailed fashion within Gartner client research.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Comments Off

Category: Uncategorized     Tags: