Windows Server 2012 is interesting for a number of reasons, but it presents something of a dilemma for those of us wishing to write about the differences between the previous Beta milestone and the current Release Candidate milestone. And that’s because Windows Server 2012, unlike its Windows 8 stable mate, was feature complete back at Beta.  

So there are no meaningful functional changes in the RC, just fit and finish work bug fixes, performance enhancements in the like. And while Microsoft has asked reviewers not to do any performance testing until the final release, since work continues in that area, the company did forward some very useful figures that nicely explain some of the meaningful performance deltas between Windows Server 2012 and its predecessor, Windows Server 2008 R2.

Logical processors

Windows Server 2012 = 320
Windows Server 2008 R2 = 64
Improvement factor = 5x

Physical memory

Windows Server 2012 = 4TB
Windows Server 2008 R2 = 1TB
Improvement factor = 4x

Virtual processors per host

Windows Server 2012 = 2048
Windows Server 2008 R2 = 512
Improvement factor = 4x

Virtual processors per virtual machine

Windows Server 2012 = 64
Windows Server 2008 R2 = 4
Improvement factor = 16x

Memory per virtual machine

Windows Server 2012 = 1TB
Windows Server 2008 R2 = 64GB
Improvement factor = 16x

Active virtual machines

Windows Server 2012 = 1024
Windows Server 2008 R2 = 384
Improvement factor = 2.7x

Cluster nodes

Windows Server 2012 = 64
Windows Server 2008 R2 = 16
Improvement factor = 4x

Cluster virtual machines

Windows Server 2012 = 4000
Windows Server 2008 R2 = 1000
Improvement factor = 4x