On Fri, Jul 1, 2011 at 8:27 AM, Andy Levy <andy.l...@gmail.com> wrote: > On Fri, Jul 1, 2011 at 07:55, Nico Kadel-Garcia <nka...@gmail.com> wrote: > >> This is more likely to happen in virtualization due to limiting the >> disk resources. Think hard about your back end disk storage. Let us >> know if you use attached storage such as a NetApp or fibre channel, >> I've got some notes on aligning those filesystems for virtualization, >> which makes a huge performance improvement. > > I'd like to see those notes if I can. We're on a NetApp device right > now but when we upgrade our environment later this year we'll be > moving storage to an Equallogic device. We have had some Subversion > performance issues with large checkouts using the NetApp device. The > Equallogic appears to be better-tuned in general (we've had no trouble > running a 100GB SQL Server DB off it) but every little bit helps.
The big booby trap was the 4096 byte block alignment, as described at http://www.citrix.com/site/resources/dynamic/partnerDocs/BestPracticesforFileSystemAlignmentinVirtualEnvironments.pdf. This is awkward to configure fat installation time for virtualized OS's, and it makes a huge performance difference. If you look up "4096 byte block NetApp Nico Kadel-Garcia" on Google, you'll see my comments on this for the last year or so. My old work accounts at Red Hat and NetApp are no longer active (I ended that contract and started a new job.), so I don't have access to the scripts I submitted on their support sites. Do you have Access to Red Hat's knowledgebase? Some notes are at https://www.redhat.com/archives/rhelv6-beta-list/2010-August/msg00069.html, and I can dig around for a published copy of the tools.
