VMWare performance resources

[websense]

Can someone help me answer these questions and/or provide me with resource(s) where I can find this information myself.

+ As you indicate, hard number comparisons are the sort of gold standard that we are looking for.+
I would be especially interested in one that did a comparison something like this:

• A. System A - VM system with two VMs, each running a server that is under load

• B. System B - A single OS, right on top of bare metal, with those same two servers running on it under the same loadTake this basic A versus B test and compare it along a bunch of parameters:

• 1. Heavy Disk I/O load

• 2. Memory intensive load

• 3. CPU intensive load

• 4. Some combinations

• 5. What did it look like at ‘startup' versus steady state

• 6. What are the boundary conditions; scenarios in each of A and B where you can cause particularly bad contention, etc.

• 7. Etc.Something like that seems to me like it would be particularly useful, as it would help to really characterize the sort of servers that could be put on VM without worry. And it would bring clarity to the sort of performance degradation (or user impact) that you should expect in various scenarios. Armed with that kind of data, it would be much easier to say "Process X can be VM'd without a second thought, while Process Y should only be VM'd under these conditions..., and Process Z will actually benefit from being put by itself on its own VM...", etc.

Thanks,

Kevin

[NTurnbull · Accepted solution]

To be honest it depends on lots of variables, but take a look at for some docs or just google vmware performance, Dell, HP and other manufacturers/software vendors have stats

Thanks,

Neil

[NTurnbull]

Oh and just saw the last bit of your post again, it's not just degredation, some apps/workload benifit from being virtualised. I'd be a bit more specific on apps as there are loads of people on the communities that would be able to lend a hand and some real world experience - not just cold hard lab tests

Thanks,

Neil