Nehalem CPU (aka Xeon 5500) and TPS on vSphere

As I wrote a while ago when you enable virtualized MMU for your virtual machine it enables Large Pages and Large Pages don’t get “TPS’ed”. The article I wrote was specifically related to AMD cause it was the only platform at the moment for which enhanced memory techniques where used. (AMD RVI!) As of vSphere 4.0 Intel EPT is also fully utilized. As expected this leads to the same “issue” as with AMD, no TPS when you enable vMMU. VMTN Community User MCWill reported this here. I wanted to specifically point this topic out to you because of the excellent replies from Kichaonline and Rajesh Venkatasubramanian. It’s worth reading the full topic if you want to get a good understanding of TPS/Virtualized MMU.

A small correction — we are currently investigating ways to fix the high memory usage issue also. Regarding TPS, as noted earlier this shoud not lead to any performance degradation. When a 2M guest memory region is backed with a machine large page, VMkernel installs page sharing hints for the 512 small (4K) pages in the region. If the system gets overcommitted at a later point, the machine large page will be broken into small pages and previously installed page sharing hints helps to quickly share the broken down small pages. So low TPS numbers when a system is undercommitted does not mean that we won’t reap benefits out of TPS when machine gets overcommitted.

http://www.yellow-bricks.com/2009/05/31/nehalem-cpu-and-tps-on-vsphere/