John,

I enjoyed your article and think it was very well-written. I do, however, disagree with the majority of your points as they are clearly stated from the perspective of tactical, rather than strategic, data center virtualization. Here is my response to your 5 points:

1. Cut prices: I’ve addressed this numerous times both on my own blog site and in other virtualization magazines for the past couple of years. The thrust is that most organizations are only partially virtualized. The reasons frequently have to do with inability to virtualize Tier-1 applications, networking concerns, lack of ISV support and politics. Virtualization, though, enables such tremendous savings that any servers remaining physical are a huge detriment to the potential ROI. VMware vSphere enables far more data center virtualization percentage than Hyper-V not only because of its superior capabilities in both performance and reliability, but also through vNetwork capabilities, superior management and flexibility. (An interesting point is that VMware is 100% virtualized internally while Microsoft's Technical Case Study released early last year shows only a 50% internal data center virtualization rate with an expectation of "virtualizing at least 80% of new servers"). You state, “Few, if any observers would claim that Microsoft's virtualization technology is better than VMware's today”. It is consequently pointless to compare pricing between vSphere and Hyper-V. The decision point should be which virtualization platform enables the scalability, manageability, reliability and security required to completely virtualize the data center. The savings enabled by utilizing vSphere make the underlying hypervisor cost delta insignificant in comparison.
2. Improve security: This is rather a perplexing point since you simply look at vSphere and not Hyper-V. Certainly vSphere is significantly more secure than Hyper-V which includes the Windows Server core and all the vulnerabilities that entails. The VMware capabilities you mention such as vShield Zones as well as others such as VMsafe and, of course, vDS, take security to a vastly higher realm than the competition. With Hyper-V, for instance, network administrators have no purview into individual virtual machines, and no way to monitor traffic over a vSwitch for compliance, auditing and troubleshooting purposes. They cannot apply network and security policies such as availability, encryption and maximum latency that follow a VM as it live-migrates between hosts.
3. Win the desktop war: This is a hard point with which to disagree. VMware is working very hard to win the war, as are many other great companies – most notably Citrix. VMware claims to have a dominating market share lead in the virtual desktop space, and certainly has tremendous traction out there. With 650M corporate PCs and laptops, it’s a huge opportunity.
4. Simplify Management: The fatal flaw in so-called multi-hypervisor management tools is that even Microsoft’s System Center Management Suite does a poor job of managing VMware vSphere; it is unlikely that other manufacturers will be able to do better. VMware focuses exclusively on optimizing virtualization resource pools of compute, storage and network. It enhances cloud attributes of agility, scalability, resiliency, accountability, security and compliance with Host Profiles, Chargeback, Orchestrator, vCloud, Site Recovery Manager, Lab Manager, Lifecycle Manager, vShield Zones, VMSafe, vApps and many other specialized management tools.
5. Don’t overhype the cloud: We recently brought 50 of our clients to San Diego for a CIO Summit. The cloud is on practically everyone’s agenda – I don’t think it’s possible to overhype it. As organizations start embracing a virtualized data center, they quickly realize they have to deal with issues such as (you point out) VM sprawl, efficiently provisioning virtual infrastructure and managing additional objects such as VMs, hosts, vSwitches, hypervisors, etc. VMware has invested, and continues to invest, a tremendous effort into developing cloud computing capabilities that optimize transition from the data center to a private cloud (and then to hybrid private/public clouds) with automated self-provisioning including metering, monitoring and chargeback.