Hi All, Has anyone out there got any impressions or have done any testing of disk IO performance in a Xen VM. I''ve done some testing in-house here and compared it to native, VMware Server and VMware ESX and come up with some results which are quite surprising. Any info anyone has on this would be great. I''m quite happy to share the results but I''d rather not say just yet for fear of introducing bias into everyone''s existing thoughts / results. Cheers *Dave Brown* IT Consultant RHCE, MCSA [image: Loftus IT] 175 Fullarton Rd Dulwich SA 5065 Ph: (08) 8304 8888 Fax: (08) 8364 2910 Mob: 0414 494 802 NOTE: This email and any files transmitted with it are confidential and may contain information intended only for the addressee(s). If you have received this communication in error, you must not copy or distribute any part of it or otherwise disclose its contents to anyone - please notify Loftus IT immediately. Loftus IT does not accept liability for any errors or omissions in the information provided herein. No representation is made that email and any files transmitted with it are virus-free - virus scanning is the responsibility of the recipient and is recommended. _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users
I am using linux based tools such as stress, iostat, bonnie and dd. The numbers will obviously not match up with the PCMark numbers but what I am looking for is how much of a disk IO performance hit occurs with Xen vs Native and VMware vs native. Steven - let me know if you still want the results. Dave On 9/20/07, Steven Hu <learn.hu@gmail.com> wrote:> > How do you do the test? When I run PCMark 2005, the score of HDD is not > stable, sometime is 2300 scores, and sometims is 3300 scores. It also fails > some times. > Couls you share your results with me? thanks. > > 2007/9/20, Dave Brown <daveb21@gmail.com>: > > > > Hi All, > > > > Has anyone out there got any impressions or have done any testing of > > disk IO performance in a Xen VM. I''ve done some testing in-house here and > > compared it to native, VMware Server and VMware ESX and come up with some > > results which are quite surprising. Any info anyone has on this would be > > great. > > > > I''m quite happy to share the results but I''d rather not say just yet for > > fear of introducing bias into everyone''s existing thoughts / results. > > > > Cheers > > > > *Dave Brown* > > IT Consultant > > RHCE, MCSA > > > > [image: Loftus IT] > > 175 Fullarton Rd > > Dulwich SA 5065 > > > > Ph: (08) 8304 8888 > > Fax: (08) 8364 2910 > > Mob: 0414 494 802 > > > > NOTE: This email and any files transmitted with it are confidential and > > may contain information intended only for the addressee(s). If you have > > received this communication in error, you must not copy or distribute any > > part of it or otherwise disclose its contents to anyone - please notify > > Loftus IT immediately. Loftus IT does not accept liability for any errors or > > omissions in the information provided herein. No representation is made that > > email and any files transmitted with it are virus-free - virus scanning is > > the responsibility of the recipient and is recommended. > > > > > > _______________________________________________ > > Xen-users mailing list > > Xen-users@lists.xensource.com > > http://lists.xensource.com/xen-users > > > >_______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users
Dear Dave, what kind of benchmark did you do? We did a Cache/Memory benchmark as well as a transfer rate benchmark to a Lustre system. Quite supprising for as was, that the CPU cache performance was quite bad but the other benchmarks showed in mean about 20% overhead. If you are interested in detailed details you can contact me. Bye Stefan Dave Brown wrote:> Hi All, > > Has anyone out there got any impressions or have done any testing of > disk IO performance in a Xen VM. I''ve done some testing in-house here > and compared it to native, VMware Server and VMware ESX and come up > with some results which are quite surprising. Any info anyone has on > this would be great. > > I''m quite happy to share the results but I''d rather not say just yet > for fear of introducing bias into everyone''s existing thoughts / results. > > Cheers > > *Dave Brown* > IT Consultant > RHCE, MCSA > > Loftus IT > 175 Fullarton Rd > Dulwich SA 5065 > > Ph: (08) 8304 8888 > Fax: (08) 8364 2910 > Mob: 0414 494 802 > > NOTE: This email and any files transmitted with it are confidential > and may contain information intended only for the addressee(s). If you > have received this communication in error, you must not copy or > distribute any part of it or otherwise disclose its contents to anyone > - please notify Loftus IT immediately. Loftus IT does not accept > liability for any errors or omissions in the information provided > herein. No representation is made that email and any files transmitted > with it are virus-free - virus scanning is the responsibility of the > recipient and is recommended. > > > ------------------------------------------------------------------------ > > _______________________________________________ > Xen-users mailing list > Xen-users@lists.xensource.com > http://lists.xensource.com/xen-users_______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users