At virtualization.info we usually don’t cover case studies. But a virtualization vendor that describes its own virtual infrastructure is well worth an exception.
In the past Microsoft already provided a few details about its internal implementation of Hyper-V, but it has now expose much more through a 23-pages whitepaper titled How Microsoft Moved to a Virtualized Infrastructure:
Microsoft.com Operations (MSCOM Ops), a team within the Microsoft Information Technology (Microsoft IT) division that manages the Microsoft.com Web site, faced the same budgetary pressures that affect many businesses today: It needed to constrain capital spending, optimize the use of existing hardware, and reduce operational costs. Using the traditional IT model of “Decommission the old and buy newer, more-powerful servers” continued to result in a lot of unused storage, network, and compute capacity. The number of physical servers provisioned to support business needs was growing at a rate of approximately 20 percent per year during a time when budget continued to shrink. In spite of earlier and significant investments in physical server hardware, Microsoft.com was using less than 10 percent of the available processing power and 30 percent of the storage space.
MSCOM Ops saw a great opportunity to address these challenges while fulfilling the Microsoft IT mission of being an early adopter of the Windows Server 2008 R2 operating system, Hyper-V, and Microsoft System Center Virtual Machine Manager 2008 R2. This paper introduces the logical design of the dynamic compute infrastructure (DCI), a virtualized hosting environment that combines these new technologies with clustering and storage area network (SAN)–based storage to provide a flexible, cost-effective hosting environment.
This paper also describes some of the automation that the team has developed to help manage Microsoft.com, and it includes operational considerations and best practices aimed at IT pros who may be considering a similar virtualization strategy. However, this paper is based on the experience and recommendations of MSCOM Ops as an early adopter. It is not intended to serve as a procedural guide. Each enterprise environment has unique circumstances; therefore, each organization should adapt the plans and lessons learned described in this paper to meet its specific needs…
Microsoft also published a 9-minutes video about this case study, available here.
It would be nice to read a similar thing from VMware and Citrix.
Thanks to Bink.nu for the news.