faz1.com blog

Now that the VMware ESX vSphere 4.0 U1 update has been released, customers are moving from 3.0 and 3.5 to 4.0 at a very accelerated pace.  U1 means that the technology is stable, the kinks have been worked out and gremlins have moved on to terrorize something else.  It is also a major mental barrier (like Service Packs in the Windows world).  Now that the barrier has been removed, there are a lot more fact and experience based analysis coming from users championing for and against new features and solutions inside of the vSphere 4 offering.

A great example of this is captured over at Search Networking comparing 2 separate articles.  The first, by Bob Plankers (lead Linux and VMware systems engineer at the University of Wisconsin-Madison, he also runs The Lone Sysadmin blog) is why the VMware vSwitch is good enough for most. The 2nd, by David Davis (a virtualization author), does a good job of articulating the why the Benefits outweigh the extra cost of Cisco Nexus 1000V.

In addition to David points, I would add one point of clarification which Bob might not be aware of. The Nexus 1000V is sold and serviced through both VMware and Cisco. In fact, VMW offers a couple of bundles of the Nexus 1000V with the vSphere Enterprise Plus licenses (both full license and upgrade license). When VMW sells the Nexus 1000V with the vSphere software, they also sell support (in conjunction with the vSphere support). Both the VMW and Cisco support teams are trained on the Nexus 1000V at the same time and both equally capable of handling support issues. And if things get really tricky, the Cisco TAC backs up VMW’s support organization with a direct line into our engineering department.

Oh, one other thing. The latest release of the Nexus 1000V software (1.2) includes a simple GUI to allow you to complete the initial config in about 7 minutes. There is a VOD posted here to show just how easy it is:

http://www.youtube.com/watch?v=-sxWiz7S-z0

It is great to see more real world analysis from real users.  Looking forward to reading more of these in the future.

Check out the Brighttalk Virtualized Data Center Webcast this Thursday….lots of good topics including mine on the Nexus 1000V!

http://www.brighttalk.com/webcasts/8134/attend

Where does policy definition, enforcement and auditing belong in a data center? Should it be silo’d for individual functions or aggregated into a centralized policy solution.  There are a log of established and emerging companies in this space (Voyence, HyTrust, Embotics, etc), but each of them focused specifically on 1 area of the DC (for the most part) like VMs, Network or Storage.  How do these solutions fit together to support APPLICATION level policy (what customers really care about) inclusive of all the pieces and parts (the building blocks, if you will) that need to come together to deliver that application

I would argue that the silo specific (VM, Network, Security, etc) policy management needs to be a tightly integrated key feature of the silo specific device management & provisioning solution (all of which should be fully automated, of course).  On top of this, there needs to be an Application level policy solution that seamlessly ties each of the silo specific offerings together (the building blocks) to allow the definition of, enforcement of and auditing of  App policy in customer/cloud environments.   The question is, how do you effectively tie all these blocks together?

What do you think?