For me,We have seem a supermicro machine,which is 2U with 2 CPU and 24 2.5 inch sata/sas drives,together with 2 onboard 10Gb Nic. I think it's good enough for both density and computing power. To another end, we are also planning to evaluating small node for ceph,say a ATOM with 2 /4 disks per node,and you could have 6 nodes in a 2U space 在 2013-3-18，6:39，"Mark Nelson" <mark.nelson@xxxxxxxxxxx> 写道： > Hi Stas, > > The SL4500 series looks like it should be a good option for large deployments, though you may want to consider going with the 2-node configuration with 25 drives each. The drive density is a bit lower but you'll have a better CPU/drive ratio and can get away with much cheaper processors (dual E5-2620s should be sufficient for 25 drives). > > It's important to keep in mind that unless you are talking about deploying multiple racks of OSDs, you are likely better off with smaller nodes with fewer drives (say 2U 12 drive boxes). That helps keep the penalty for losing a node from being too dramatic. > > Both the SL4500 and the Dell C8000 allow you to have configurations with multiple nodes in 1 chassis with fewer drives, so they are kind of an interesting compromise between high density and keeping the drives-per-node count lower. Granted, they both tend to be more expensive than supermicro gear, so like always it's a giant balancing act. :) > > Mark > > On 03/17/2013 04:31 PM, Stas Oskin wrote: >> Hi. >> >> First of all, nice to meet you, and thanks for the great software! >> >> I've thoroughly read the benchmarks on the SuperMicro hardware with and >> without SSD combinations, and wondered if there were any tests done on >> HP file server. >> >> According to this article: >> http://www.theregister.co.uk/2012/11/15/hp_proliant_sl4500_big_data_servers/ >> >> This server in single node configuration is ideal for clustered systems >> (OpenStack in this case), holds 60 3.5 drives and can push up to 1M >> IOPS. Being priced as $7,643, it seems to make a serious competition to >> SuperMicro's hardware. >> >> Any idea what throughput can be achieved on this machine with Ceph? >> >> Regards. >> >> >> >> _______________________________________________ >> ceph-users mailing list >> ceph-users@xxxxxxxxxxxxxx >> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com >> > > _______________________________________________ > ceph-users mailing list > ceph-users@xxxxxxxxxxxxxx > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |