On Tue, Mar 15, 2022 at 7:43 AM Jonas Liechti <[email protected]> wrote: > Thank you for the confirmation Strahil! > > As our current environment is more or less the same (except the Hardware > RAID, which is not possible with NVMe disks), we planned to use Gluster. I > guess we will proceed as we originally planned as we are satisfied with the > performance. > Sorry for being late to the party.
Two items to consider: A. If you have more than two drives, consider using MDRAID 10 on NVME drives We have been using it for a while now and getting *very* good performance (more or less in-line with a single NVME drive). As an added bonus, we managed to survive a single drive failure without fault. B .If possible use a 40GbE network between the nodes instead of using 10GbE. It makes a huge impact. (SSDs or even a large HDD array can easily saturate a single 10GbE link). - Gilboa > On 3/11/22 07:08, Strahil Nikolov via Users wrote: > > Red Hat Gluster Storage is discontinued, but the Gluster (upstream) is > pretty active and as Sandro Bonazzola (RH) confirmed -> there are no plans > to remove support for Gluster. > I think it's still a good choice, especially if you don't have SAN/ > Higly-Available NFS. > > Also, storage migration is transparent for the VMs, so you can add SAN on > a later stage and move all VMs from Gluster to SAN without disruption* . > > Keep in mind that Gluster is a tier2 storage and if you really need a lot > of IOPS, CEPH might be suitable. > > > Best Regards, > Strahil Nikolov > > *: Note that this is valid if the FUSE client is used. Other oVirt users > report huge performance increase when using libgfapi interface, but it has > drawbacks like storage migration can happen only when you switch off > libgfapi, power off the VM (on a scheduled basis), power on the VM, live > migrate the VM to other storage type, enable libgfapi again for the rest of > the VMs. > > > > Thanks to Nikolov and Strahil for the valuable input! I was off for a few > weeks, so I would like to apologize if I'm potentially reviving a zombie > thread. > > I am a bit confused about where to go with this environment after the > discontinuation of the hyperconverged setup. What alternative options are > there for us? Or do you think going the Gluster way would still be > advisable, even though it seems as it is being discontinued over time? > > Thanks for any input on this! > > Best regards, > Jonas > On 1/22/22 14:31, Strahil Nikolov via Users wrote: > > Using the wizzard is utilizing the Gluster Andible roles. > I would highly recommend using it, unless you know what you are doing (for > example storage alignment when using Hardware raid). > > Keep in mind that the DHT xlator (the logic in distributed volumes) is > shard aware, so your shards are spread between subvolumes and additional > performance can be gained.So using replicated-distributed volumes have > their benefits. > > If you decide to avoid the software raid, use only replica3 volumes as > with SSDs/NVMEs usually the failures are not physical, but logical (maximum > writes reached -> predictive failure -> total failure). > > Also, consider mounting via noatime/relatime and > context="system_u:object_r:glusterd_brick_t:s0" for your gluster bricks. > > Best Regards, > Strahil Nikolov > > On Fri, Jan 21, 2022 at 11:00, Gilboa Davara > <[email protected]> <[email protected]> wrote: > _______________________________________________ > Users mailing list -- [email protected] > To unsubscribe send an email to [email protected] > Privacy Statement: https://www.ovirt.org/privacy-policy.html > oVirt Code of Conduct: > https://www.ovirt.org/community/about/community-guidelines/ > List Archives: > > https://lists.ovirt.org/archives/list/[email protected]/message/U2ZEWLRF5D6FENQEI5QXL77CMWB7XF32/ > > _______________________________________________ > Users mailing list -- [email protected] > To unsubscribe send an email to [email protected] > Privacy Statement: https://www.ovirt.org/privacy-policy.html > oVirt Code of Conduct: > https://www.ovirt.org/community/about/community-guidelines/ > List Archives: > https://lists.ovirt.org/archives/list/[email protected]/message/26AHNDSOJSIVTGYOEUFOY444YYBZCAIW/ > > _______________________________________________ > Users mailing list -- [email protected] > To unsubscribe send an email to [email protected] > Privacy Statement: https://www.ovirt.org/privacy-policy.html > oVirt Code of Conduct: > https://www.ovirt.org/community/about/community-guidelines/ > List Archives: > https://lists.ovirt.org/archives/list/[email protected]/message/JFFS2T25TIIHFEQMS2Y3BU4DARSIDE3U/ >
_______________________________________________ Users mailing list -- [email protected] To unsubscribe send an email to [email protected] Privacy Statement: https://www.ovirt.org/privacy-policy.html oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/[email protected]/message/DWLJDJ5VO2CV6DNYLMVOWSUXKCQQLPUW/
