Hi all,
I've been doing some tests on guest vms with different cpu options, quantities and back end storage options. It's been rather interesting so far, I'm not quite done yet (still have to test GlusterFS both through native gluster mapping and NFS to proxmox and NFS to Xenserver) - in case anyone is interested in following the information, I have it in a spreadsheet here:
https://docs.google.com/spreadsheets/d/1h_zgVgcsaUor1aAFUuwkfLgtcl4i9QpYdmzlX_bKedc/edit?usp=sharing
Right now they are all single runs of the benchmark software, eventually when I get enough time I'll probably do multiple runs to get averages together.
If anyone has any other tests they can think of that might be beneficial to try, let me know. Note that this is also using a 32bit guest OS to match the majority of VMs we currently have in VMWare that we will migrate out to our (I should say my as I make the final decision on this one at work) end use system, but I'll probably test with a 64bit OS at some point as well.
I've been doing some tests on guest vms with different cpu options, quantities and back end storage options. It's been rather interesting so far, I'm not quite done yet (still have to test GlusterFS both through native gluster mapping and NFS to proxmox and NFS to Xenserver) - in case anyone is interested in following the information, I have it in a spreadsheet here:
https://docs.google.com/spreadsheets/d/1h_zgVgcsaUor1aAFUuwkfLgtcl4i9QpYdmzlX_bKedc/edit?usp=sharing
Right now they are all single runs of the benchmark software, eventually when I get enough time I'll probably do multiple runs to get averages together.
If anyone has any other tests they can think of that might be beneficial to try, let me know. Note that this is also using a 32bit guest OS to match the majority of VMs we currently have in VMWare that we will migrate out to our (I should say my as I make the final decision on this one at work) end use system, but I'll probably test with a 64bit OS at some point as well.