Hello, As discussed on this very ML and by looking at the moderately well hidden full specifications, using any 7xx Intel SSD for journals is a fools errand. Their endurance is far too low in comparison to their IOPS capacity. Regards, Chibi On Thu, 11 Feb 2016 15:15:12 +0000 Alex Leake wrote: > Michael, > > > I wouldn't be concerned with SAS expanders - so long as you've got > enough bandwidth to the HBA / RAID controller? > > > The main consideration with the SSDs is the ratio to disk. When you > loose an SSD all the OSDs journalling to that will be inconsistent, > effectively off-lining them. Too many disks can just hammer the SSD > making it the bottleneck. > > > Even if the intel 720 SSDs can provide exceptional performance, more > than 8 seems like it could be risky / impact performance? > > > You could consider cache tiers instead, make two separate pools, one > comprising of SSD the other just of disk then flush cold objects from > the SSDs. > > > http://docs.ceph.com/docs/master/rados/operations/cache-tiering/ > > > > Kind Regards, > > Alex. > > ________________________________ > From: Michael Barkdoll <mabarkdoll@xxxxxxxxx> > Sent: 11 February 2016 15:05 > To: Alex Leake > Cc: ceph-users@xxxxxxxxxxxxxx > Subject: Re: Dell Ceph Hardware recommendations > > Thanks Alex, > > Dell provided me with their current recommended cloud architecture for a > redhat openstack/ceph deployment. > http://ja.community.dell.com/techcenter/m/mediagallery/3693/download > > The R730XD is also recommended by Dell. I'd save about $2000 per > server, if I purchase the Dell PowerEdge T630 compared to the R730XD. > The RAID controllers are the same, I'm just curious if there would be > any issue with the SAS expansion backplate. An old InkTank deployment > guide from 2013 recommended against using SAS expansion backplates. > Yet, RAID controllers are now 12Gb so I'm not certain if this is still > an issue? > > Also, I was thinking about using a Intel 750 Series AIC 800GB > PCI-Express 3.0 x4 MLC Internal Solid State Drive (SSD) SSDPEDMW800G4X1 > 800GB model rather than the 1.2TB model for caching. Does anyone know > how many hard disks I can use with it? With the 1.2TB model, I may have > read some where that it supports around 12-16 disks. > > I'm still leaning toward using the T630 with maybe a PERC H730 for cost > savings. I'm curious if the PERC H730 offers any advantage of the PERC > H330 for Ceph? Thanks so much everyone for the much needed feedback! > > > Michael Barkdoll > > On Thu, Feb 11, 2016 at 7:54 AM, Alex Leake > <A.M.D.Leake@xxxxxxxxxx<mailto:A.M.D.Leake@xxxxxxxxxx>> wrote: Hello > Michael?, > > I maintain a small Ceph cluster at the University of Bath, our cluster > consists of: > > Monitors: > 3 x Dell PowerEdge R630 > > - 2x Intel(R) Xeon(R) CPU E5-2609 v3 > - 64GB RAM > - 4x 300GB SAS (RAID 10) > > > OSD Nodes: > 6 x Dell PowerEdge R730XD & MD1400 Shelves > > - 2x Intel(R) Xeon(R) CPU E5-2650 > - 128GB RAM > - 2x 600GB SAS (OS - RAID1) > - 2x 200GB SSD (PERC H730) > - 14x 6TB NL-SAS (PERC H730) > - 12x 4TB NL-SAS (PERC H830 - MD1400) > > > Please let me know if you want any more info. > > In my experience thus far, I've found this ratio is not useful for cache > tiering etc - the SSDs are in a separate pool. > > If I could start over, I'd go for fewer OSDs / host - and no SSDs (or a > much better ratio - like 4:1). > > > Kind Regards, > Alex. > -- Christian Balzer Network/Systems Engineer chibi@xxxxxxx Global OnLine Japan/Rakuten Communications http://www.gol.com/ _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |