VMware Cloud Community
IT_Architect
Enthusiast
Enthusiast

Where can I find VMark Benchmarks?

Where can I find VMark Benchmarks?  I can't find much except a few from the major manufacturers where there are no two systems alike, and they are all huge data center machine.  I need to compare the 5600 vs the 2600 lines for performance in VMware and web servers.  I understand very well the differences between the two architectures, but that is useless information for a value analysis.  While clock-for-clock the 2600s are faster, most CPU ratings are significantly influenced by the 2600's AVT instruction set, which is unusable for most server applications.  I've determined that for LAMP stack servers, the 2600s clock-for-clock have at least a 14% advantage, but most certainly not more than a 16% advantage.  The missing element in my value analysis is the performance impact the 2600 architecture has on ESXi 5.1.  "It should be...", or "In theory...", or "I noticed they are a lot faster..." is useless drivel.  I need data from benchmarks that I can plug into a calculator.

Thanks!

0 Kudos
2 Replies
AndySimmons
Hot Shot
Hot Shot

I'm not sure if they'd be of much use to you, but we'll be doing a compute bake-off in the next couple of months, and I'd be happy to share the VMark results. Our existing vSphere hosts are HP BL460 G7s with x5670s. We have systems from HP, IBM, and Cisco on the way, all with the e5-2665. We're more interested in the blade management than anything, but if the VMark results would be helpful, let me know and I'd be happy to share.

Based on your first post, I'm assuming you've already seen this and this, but I'll include them here in case anyone else stumbles across this.

-Andy VCAP5-DCA, VCP-DV 4/5, MCSE, space camp graduate.
0 Kudos
IT_Architect
Enthusiast
Enthusiast

I'll probably pull a trigger today or tommorrow.  I'm just going to call it 16%.  My logic is that's the difference it makes in normal programs that don't use the AVX extensions.

Thanks,
Jack

0 Kudos