Hello, On Thu, 08 May 2014 06:33:51 +0200 (CEST) Alexandre DERUMIER wrote:
> Hi Christian, > > Do you have tried without raid6, to have more osd ? No and that is neither an option nor the reason for any performance issues here. If you re-read my original mail it clearly states that the same fio can achieve 110000 IOPS on that raid and that it is not busy at all during the test. > (how many disks do you have begin the raid6 ?) > 11 per OSD. This will affect the amount of sustainable IOPS of course, but in this test case every last bit should (and does) fit into the caches. From the RBD client the transaction should be finished once the primary and secondary OSD for the PG in question have ACK'ed things. > > Aslo, I known that direct ios can be quite slow with ceph, > > maybe can you try without --direct=1 > I can, but that is not the test case here. For the record that pushes it to 12k IOPS, with the journal SSDs reaching about 30% utilization and the actual OSDs up to 5%. So much better, but still quite some capacity for improvement. > and also enable rbd_cache > > ceph.conf > [client] > rbd cache = true > I have that set of course, as well as specifically "writeback" for the KVM instance in question. Interestingly I see no difference at all with a KVM instance that is set explicitly to "none", but that's not part of this particular inquiry either. Christian > > > > ----- Mail original ----- > > De: "Christian Balzer" <ch...@gol.com> > À: "Gregory Farnum" <g...@inktank.com>, ceph-users@lists.ceph.com > Envoyé: Jeudi 8 Mai 2014 04:49:16 > Objet: Re: [ceph-users] Slow IOPS on RBD compared to journal and backing > devices > > On Wed, 7 May 2014 18:37:48 -0700 Gregory Farnum wrote: > > > On Wed, May 7, 2014 at 5:57 PM, Christian Balzer <ch...@gol.com> > > wrote: > > > > > > Hello, > > > > > > ceph 0.72 on Debian Jessie, 2 storage nodes with 2 OSDs each. The > > > journals are on (separate) DC 3700s, the actual OSDs are RAID6 > > > behind an Areca 1882 with 4GB of cache. > > > > > > Running this fio: > > > > > > fio --size=400m --ioengine=libaio --invalidate=1 --direct=1 > > > --numjobs=1 --rw=randwrite --name=fiojob --blocksize=4k > > > --iodepth=128 > > > > > > results in: > > > > > > 30k IOPS on the journal SSD (as expected) > > > 110k IOPS on the OSD (it fits neatly into the cache, no surprise > > > there) 3200 IOPS from a VM using userspace RBD > > > 2900 IOPS from a host kernelspace mounted RBD > > > > > > When running the fio from the VM RBD the utilization of the journals > > > is about 20% (2400 IOPS) and the OSDs are bored at 2% (1500 IOPS > > > after some obvious merging). > > > The OSD processes are quite busy, reading well over 200% on atop, > > > but the system is not CPU or otherwise resource starved at that > > > moment. > > > > > > Running multiple instances of this test from several VMs on > > > different hosts changes nothing, as in the aggregated IOPS for the > > > whole cluster will still be around 3200 IOPS. > > > > > > Now clearly RBD has to deal with latency here, but the network is > > > IPoIB with the associated low latency and the journal SSDs are the > > > (consistently) fasted ones around. > > > > > > I guess what I am wondering about is if this is normal and to be > > > expected or if not where all that potential performance got lost. > > > > Hmm, with 128 IOs at a time (I believe I'm reading that correctly?) > Yes, but going down to 32 doesn't change things one iota. > Also note the multiple instances I mention up there, so that would be > 256 IOs at a time, coming from different hosts over different links and > nothing changes. > > > that's about 40ms of latency per op (for userspace RBD), which seems > > awfully long. You should check what your client-side objecter settings > > are; it might be limiting you to fewer outstanding ops than that. > > Googling for client-side objecter gives a few hits on ceph devel and > bugs and nothing at all as far as configuration options are concerned. > Care to enlighten me where one can find those? > > Also note the kernelspace (3.13 if it matters) speed, which is very much > in the same (junior league) ballpark. > > > If > > it's available to you, testing with Firefly or even master would be > > interesting — there's some performance work that should reduce > > latencies. > > > Not an option, this is going into production next week. > > > But a well-tuned (or even default-tuned, I thought) Ceph cluster > > certainly doesn't require 40ms/op, so you should probably run a wider > > array of experiments to try and figure out where it's coming from. > > I think we can rule out the network, NPtcp gives me: > --- > 56: 4096 bytes 1546 times --> 979.22 Mbps in 31.91 usec > --- > > For comparison at about 512KB it reaches maximum throughput and still > isn't that laggy: > --- > 98: 524288 bytes 121 times --> 9700.57 Mbps in 412.35 usec > --- > > So with the network performing as well as my lengthy experience with > IPoIB led me to believe, what else is there to look at? > The storage nodes perform just as expected, indicated by the local fio > tests. > > That pretty much leaves only Ceph/RBD to look at and I'm not really sure > what experiments I should run on that. ^o^ > > Regards, > > Christian > > > -Greg > > Software Engineer #42 @ http://inktank.com | http://ceph.com > > > > -- Christian Balzer Network/Systems Engineer ch...@gol.com Global OnLine Japan/Fusion Communications http://www.gol.com/ _______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
