Re: New cluster performance analysis

[Jan Schermer <jan@xxxxxxxxxxx>]

On 04 Dec 2015, at 14:31, Adrien Gillard <gillard.adrien@xxxxxxxxx> wrote:

After some more tests :

 - The pool being used as cache pool has no impact on performance, I get the same results with a "dedicated" replicated pool.

 - You are right Jan, on raw devices I get better performance on a volume if I fill it first, or at least if I write a zone that already has been allocated

 - The same seem to apply when the test is run on the mounted filesystem.

Yeah. The the first (raw device) is because the objects on OSDs get "thick" in the process.

The second (filesystem) is because of both the OSD objects getting thick and the guest filesystem getting thick.

Preallocating the space can speed up things considerably (like 100x)).

Unfortunately I haven't found a way to convince fallocate() &co. to thick provision files.

Jan

On Thu, Dec 3, 2015 at 2:49 PM, Adrien Gillard <gillard.adrien@xxxxxxxxx> wrote:

I did some more tests : 

fio on a raw RBD volume (4K, numjob=32, QD=1) gives me around 3000 IOPS

I also tuned xfs mount options on client (I realized I didn't do that already) and with "largeio,inode64,swalloc,logbufs=8,logbsize=256k,attr2,auto,nodev,noatime,nodiratime" I get better performance :

4k-32-1-randwrite-libaio: (groupid=0, jobs=32): err= 0: pid=26793: Thu Dec  3 10:45:55 2015

  write: io=1685.3MB, bw=5720.1KB/s, iops=1430, runt=301652msec

    slat (usec): min=5, max=1620, avg=41.61, stdev=25.82

    clat (msec): min=1, max=4141, avg=14.61, stdev=112.55

     lat (msec): min=1, max=4141, avg=14.65, stdev=112.55

    clat percentiles (msec):

     |  1.00th=[    3],  5.00th=[    4], 10.00th=[    4], 20.00th=[    4],

     | 30.00th=[    4], 40.00th=[    5], 50.00th=[    5], 60.00th=[    5],

     | 70.00th=[    5], 80.00th=[    6], 90.00th=[    7], 95.00th=[    7],

     | 99.00th=[  227], 99.50th=[  717], 99.90th=[ 1844], 99.95th=[ 2245],

     | 99.99th=[ 3097]

So, more than 50% improvement but it actually varies quite a lot between tests (sometimes I get a bit more than 1000). If I run the test fo 30 minutes it drops to 900 IOPS.

As you suggested I also filled a volume with zeros (dd if=/dev/zero of=/dev/rbd1 bs=1M) and then ran fio on the raw device, I didn't see a lot of improvement.

If I run fio test directly on block devices I seem to saturate the spinners, [1] is a graph of IO load on one of the OSD host. 

[2] is the same OSD graph but when the test is done on a device mounted and formatted with XFS on the client. 

If I get half of the IOPS on the XFS volume because of the journal, shouldn't I get the same amount of IOPS on the backend ? 

[3] shows what happen if I run the test for 30 minutes.

During the fio tests on the raw device, load average on the OSD servers increases up to 13/14 and I get a bit of iowait (I guess because the OSD are busy)

During the fio tests on the raw device, load average on the OSD servers peaks at the beginning and decreases to 5/6, but goes trough the roof on the client.

Scheduler is deadline for all the drives, I didn't try to change it yet.

What I don't understand, even with your explanations, are the rados results. From what I understand it performs at the RADOS level and thus should not be impacted by client filesystem.

Given the results above I guess you are right and this has to do with the client filesystem.

The cluster will be used for backups, write IO size during backups is around 150/200K (I guess mostly sequential) and I am looking for the highest bandwith and parallelization.

@Nick, I will try to create a new stand alone replicated pool.

[1] http://postimg.org/image/qvtvdq1n1/

[2] http://postimg.org/image/nhf6lzwgl/

[3] http://postimg.org/image/h7l0obw7h/

On Thu, Dec 3, 2015 at 1:30 PM, Nick Fisk <nick@xxxxxxxxxx> wrote:

Couple of things to check

1.      Can you create just a normal non cached pool and test performance to rule out any funnies going on there.

2.      Can you also run something like iostat during the benchmarks and see if it looks like all your disks are getting saturated.

_____________________________________________
From: ceph-users [mailto:ceph-users-bounces@xxxxxxxxxxxxxx] On Behalf Of Adrien Gillard
Sent: 02 December 2015 21:33
To: ceph-users@xxxxxxxx
Subject: New cluster performance analysis




Hi everyone,

 


I am currently testing our new cluster and I would like some feedback on the numbers I am getting.

 


For the hardware :

7 x OSD : 2 x Intel 2640v3 (8x2.6GHz), 64B RAM, 2x10Gbits LACP for public net., 2x10Gbits LACP for cluster net., MTU 9000

1 x MON : 2 x Intel 2630L (6x2GHz), 32GB RAM and Intel DC SSD, 2x10Gbits LACP for public net., MTU 9000

2 x MON : VMs (8 cores, 8GB RAM), backed by SSD

 


Journals are 20GB partitions on SSD

 


The system is CentOS 7.1 with stock kernel (3.10.0-229.20.1.el7.x86_64). No particular system optimizations.

 


Ceph is Infernalis from Ceph repository  : ceph version 9.2.0 (bb2ecea240f3a1d525bcb35670cb07bd1f0ca299)

 


[cephadm@cph-adm-01  ~/scripts]$ ceph -s

    cluster 259f65a3-d6c8-4c90-a9c2-71d4c3c55cce

     health HEALTH_OK

     monmap e1: 3 mons at {clb-cph-frpar1-mon-02=x.x.x.2:6789/0,clb-cph-frpar2-mon-01=x.x.x.1:6789/0,clb-cph-frpar2-mon-03=x.x.x.3:6789/0}

            election epoch 62, quorum 0,1,2 clb-cph-frpar2-mon-01,clb-cph-frpar1-mon-02,clb-cph-frpar2-mon-03

     osdmap e844: 84 osds: 84 up, 84 in

            flags sortbitwise

      pgmap v111655: 3136 pgs, 3 pools, 3166 GB data, 19220 kobjects

            8308 GB used, 297 TB / 305 TB avail

                3136 active+clean

 


My ceph.conf :

 


[global]

fsid = 259f65a3-d6c8-4c90-a9c2-71d4c3c55cce

mon_initial_members = clb-cph-frpar2-mon-01, clb-cph-frpar1-mon-02, clb-cph-frpar2-mon-03

mon_host = x.x.x.1,x.x.x.2,x.x.x.3

auth_cluster_required = cephx

auth_service_required = cephx

auth_client_required = cephx

filestore_xattr_use_omap = true

public network = 10.25.25.0/24

cluster network = 10.25.26.0/24

debug_lockdep = 0/0

debug_context = 0/0

debug_crush = 0/0

debug_buffer = 0/0

debug_timer = 0/0

debug_filer = 0/0

debug_objecter = 0/0

debug_rados = 0/0

debug_rbd = 0/0

debug_journaler = 0/0

debug_objectcatcher = 0/0

debug_client = 0/0

debug_osd = 0/0

debug_optracker = 0/0

debug_objclass = 0/0

debug_filestore = 0/0

debug_journal = 0/0

debug_ms = 0/0

debug_monc = 0/0

debug_tp = 0/0

debug_auth = 0/0

debug_finisher = 0/0

debug_heartbeatmap = 0/0

debug_perfcounter = 0/0

debug_asok = 0/0

debug_throttle = 0/0

debug_mon = 0/0

debug_paxos = 0/0

debug_rgw = 0/0

 


[osd]

osd journal size = 0

osd mount options xfs = "rw,noatime,inode64,logbufs=8,logbsize=256k"

filestore min sync interval = 5

filestore max sync interval = 15

filestore queue max ops = 2048

filestore queue max bytes = 1048576000

filestore queue committing max ops = 4096

filestore queue committing max bytes = 1048576000

filestore op thread = 32

filestore journal writeahead = true

filestore merge threshold = 40

filestore split multiple = 8

 


journal max write bytes = 1048576000

journal max write entries = 4096

journal queue max ops = 8092

journal queue max bytes = 1048576000

 


osd max write size = 512

osd op threads = 16

osd disk threads = 2

osd op num threads per shard = 3

osd op num shards = 10

osd map cache size = 1024

osd max backfills = 1

osd recovery max active = 2

 


I have set up 2 pools : one for cache with 3x replication in front of an EC pool. At the moment I am only interested in the cache pool, so no promotions/flushes/evictions happen.

(I know, I am using the same set of OSD for hot and cold data, but in my use case they should not be used at the same time.)

 


I am accessing the cluster via RBD volumes mapped with the kernel module on CentOS 7.1. These volumes are formatted in XFS on the clients.

 


The journal SSDs seem to perform quite well according to the results of Sebastien Han’s benchmark suggestion (they are Sandisk) :

write: io=22336MB, bw=381194KB/s, iops=95298, runt= 60001msec (this is for numjob=10)

 


Here are the rados bench tests :

 


rados bench -p rbdcache 120 write -b 4K -t 32 --no-cleanup

Total time run:         121.410763

Total writes made:      65357

Write size:             4096

Bandwidth (MB/sec):     2.1

Stddev Bandwidth:       0.597

Max bandwidth (MB/sec): 3.89

Min bandwidth (MB/sec): 0.00781

Average IOPS:           538

Stddev IOPS:            152

Max IOPS:               995

Min IOPS:               2

Average Latency:        0.0594

Stddev Latency:         0.18

Max latency:            2.82

Min latency:            0.00494

 


And the results of the fio test with the following parameters :

 


[global]

size=8G

runtime=300

ioengine=libaio

invalidate=1

direct=1

sync=1

fsync=1

numjobs=32

rw=randwrite

name=4k-32-1-randwrite-libaio

blocksize=4K

iodepth=1

directory=/mnt/rbd

group_reporting=1

4k-32-1-randwrite-libaio: (groupid=0, jobs=32): err= 0: pid=20442: Wed Dec  2 21:38:30 2015

  write: io=992.11MB, bw=3389.3KB/s, iops=847, runt=300011msec

    slat (usec): min=5, max=4726, avg=40.32, stdev=41.28

    clat (msec): min=2, max=2208, avg=19.35, stdev=74.34

     lat (msec): min=2, max=2208, avg=19.39, stdev=74.34

    clat percentiles (msec):

     |  1.00th=[    3],  5.00th=[    4], 10.00th=[    4], 20.00th=[    4],

     | 30.00th=[    4], 40.00th=[    5], 50.00th=[    5], 60.00th=[    5],

     | 70.00th=[    6], 80.00th=[    7], 90.00th=[   38], 95.00th=[   63],

     | 99.00th=[  322], 99.50th=[  570], 99.90th=[ 1074], 99.95th=[ 1221],

     | 99.99th=[ 1532]

    bw (KB  /s): min=    1, max=  448, per=3.64%, avg=123.48, stdev=102.09

    lat (msec) : 4=30.30%, 10=51.27%, 20=1.71%, 50=9.91%, 100=4.03%

    lat (msec) : 250=1.55%, 500=0.62%, 750=0.33%, 1000=0.16%

  cpu          : usr=0.09%, sys=0.25%, ctx=963114, majf=0, minf=928

  IO depths    : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%

     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%

     complete  : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%

     issued    : total=r=0/w=254206/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0

     latency   : target=0, window=0, percentile=100.00%, depth=1

Run status group 0 (all jobs):

  WRITE: io=992.11MB, aggrb=3389KB/s, minb=3389KB/s, maxb=3389KB/s, mint=300011msec, maxt=300011msec

Disk stats (read/write):

  rbd0: ios=0/320813, merge=0/10001, ticks=0/5670847, in_queue=5677825, util=100.00%







And a job closer to what the actual workload would be (blocksize=200K, numjob=16, QD=32)

200k-16-32-randwrite-libaio: (groupid=0, jobs=16): err= 0: pid=4828: Wed Dec  2 18:58:53 2015

  write: io=47305MB, bw=161367KB/s, iops=806, runt=300189msec

    slat (usec): min=17, max=358430, avg=155.11, stdev=2361.49

    clat (msec): min=9, max=3584, avg=613.88, stdev=168.68

     lat (msec): min=10, max=3584, avg=614.04, stdev=168.66

    clat percentiles (msec):

     |  1.00th=[  375],  5.00th=[  469], 10.00th=[  502], 20.00th=[  537],

     | 30.00th=[  553], 40.00th=[  578], 50.00th=[  594], 60.00th=[  603],

     | 70.00th=[  627], 80.00th=[  652], 90.00th=[  701], 95.00th=[  881],

     | 99.00th=[ 1205], 99.50th=[ 1483], 99.90th=[ 2638], 99.95th=[ 2671],

     | 99.99th=[ 2999]

    bw (KB  /s): min=  260, max=18181, per=6.31%, avg=10189.40, stdev=2009.86

    lat (msec) : 10=0.01%, 20=0.01%, 50=0.01%, 100=0.02%, 250=0.08%

    lat (msec) : 500=9.26%, 750=83.21%, 1000=4.09%

  cpu          : usr=0.22%, sys=0.55%, ctx=719279, majf=0, minf=433

  IO depths    : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=99.8%, >=64=0.0%

     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%

     complete  : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.1%, 64=0.0%, >=64=0.0%

     issued    : total=r=0/w=242203/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0

     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):

  WRITE: io=47305MB, aggrb=161367KB/s, minb=161367KB/s, maxb=161367KB/s, mint=300189msec, maxt=300189msec

Disk stats (read/write):

  rbd0: ios=1/287809, merge=0/18393, ticks=50/5887593, in_queue=5887504, util=99.91%




The 4k block performance does not interest me so much but is given as a reference. I am more looking for throughput, but anyway, the numbers seem quite low.

Let's take IOPS, assuming the spinners can do 50 (4k) synced sustained IOPS (I hope they can do more ^^), we should be around 50x84/3 = 1400 IOPS, which is far from rados bench (538) and fio (847). And surprisingly fio numbers are greater than rados.

So I don't know wether I am missing something here or if something is going wrong (maybe both !).

Any input would be very valuable.

Thank you,

Adrien << File: ATT00001.txt >>

--

-----------------------------------------------------------------------------------------
Adrien GILLARD

+33 (0)6 29 06 16 31
gillard.adrien@xxxxxxxxx

--

-----------------------------------------------------------------------------------------
Adrien GILLARD

+33 (0)6 29 06 16 31
gillard.adrien@xxxxxxxxx

_______________________________________________
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