some tests with fio ioengine libaio and psync

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi:

I have a ceph cluster with 19 ssds as cache tier. Replication is 3. 100+ osd as backend storage.

I make some performance  tests use fio  follow those steps

 

rbd create liuliangtest -p vms size 20G

rbd map -p vms liuliangtest

mkfs.ext4 /dev/rbd0

mount /dev/rbd0 /mnt

    The test  command and results are as follows. My  doubt is  with different ioengine. Other options are same.  The iops with libaio  is 7971 but psync is only 299.  Is this correct why

 

[root@iaas01 mnt]# fio --bs=4k --ioengine=libaio --direct=1 --rw=randwrite --runtime=60 --group_reporting --name=fio-write --size=1G --filename=/mnt/test.db -numjobs=32 --iodepth=1

 

fio-write: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=1

...

fio-3.1

Starting 32 processes

Jobs: 32 (f=32): [w(32)][100.0%][r=0KiB/s,w=30.2MiB/s][r=0,w=7731 IOPS][eta 00m:00s]

fio-write: (groupid=0, jobs=32): err= 0: pid=3357231: Tue Dec 17 22:06:11 2019

  write: IOPS=7971, BW=31.1MiB/s (32.7MB/s)(1869MiB/60012msec)

    slat (usec): min=5, max=5723, avg=17.87, stdev=39.82

    clat (usec): min=1149, max=428306, avg=3991.69, stdev=9606.04

     lat (usec): min=1161, max=428320, avg=4009.88, stdev=9606.36

    clat percentiles (usec):

     |  1.00th=[  1500],  5.00th=[  1663], 10.00th=[  1778], 20.00th=[  1942],

     | 30.00th=[  2089], 40.00th=[  2245], 50.00th=[  2409], 60.00th=[  2671],

     | 70.00th=[  3032], 80.00th=[  3851], 90.00th=[  6259], 95.00th=[  9503],

     | 99.00th=[ 23462], 99.50th=[ 39060], 99.90th=[160433], 99.95th=[185598],

     | 99.99th=[256902]

   bw (  KiB/s): min=  200, max= 1595, per=3.13%, avg=997.40, stdev=318.73, samples=3840

   iops        : min=   50, max=  398, avg=249.17, stdev=79.62, samples=3840

  lat (msec)   : 2=23.70%, 4=57.33%, 10=14.40%, 20=3.32%, 50=0.88%

  lat (msec)   : 100=0.13%, 250=0.25%, 500=0.01%

  cpu          : usr=0.19%, sys=0.57%, ctx=480565, majf=0, minf=986

  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 rwt: total=0,478401,0, short=0,0,0, dropped=0,0,0

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

 

Run status group 0 (all jobs):

  WRITE: bw=31.1MiB/s (32.7MB/s), 31.1MiB/s-31.1MiB/s (32.7MB/s-32.7MB/s), io=1869MiB (1960MB), run=60012-60012msec

 

Disk stats (read/write):

  rbd0: ios=0/477511, merge=0/11, ticks=0/1902034, in_queue=1903400, util=99.90%

 

 

[root@iaas01 mnt]# fio --bs=4k --ioengine=psync --direct=1 --rw=randwrite --runtime=60 --group_reporting --name=fio-write --size=1G --filename=/mnt/test.db -numjobs=32 --iodepth=1

 

fio-write: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=psync, iodepth=1

...

fio-3.1

Starting 32 processes

Jobs: 32 (f=32): [w(32)][100.0%][r=0KiB/s,w=1001KiB/s][r=0,w=250 IOPS][eta 00m:00s]

fio-write: (groupid=0, jobs=32): err= 0: pid=3361544: Tue Dec 17 22:08:55 2019

  write: IOPS=299, BW=1200KiB/s (1229kB/s)(70.5MiB/60168msec)

    clat (usec): min=1583, max=426473, avg=106536.74, stdev=58040.24

     lat (usec): min=1583, max=426474, avg=106537.41, stdev=58040.14

    clat percentiles (usec):

     |  1.00th=[  1926],  5.00th=[  2278], 10.00th=[  3097], 20.00th=[ 86508],

    | 30.00th=[ 93848], 40.00th=[ 99091], 50.00th=[104334], 60.00th=[110625],

     | 70.00th=[117965], 80.00th=[130548], 90.00th=[158335], 95.00th=[208667],

     | 99.00th=[295699], 99.50th=[350225], 99.90th=[404751], 99.95th=[413139],

     | 99.99th=[425722]

   bw (  KiB/s): min=    8, max=  112, per=3.13%, avg=37.54, stdev=11.62, samples=3840

   iops        : min=    2, max=   28, avg= 9.38, stdev= 2.90, samples=3840

  lat (msec)   : 2=1.66%, 4=9.32%, 10=1.32%, 20=0.34%, 50=0.04%

  lat (msec)   : 100=28.89%, 250=55.22%, 500=3.21%

  cpu          : usr=0.01%, sys=0.06%, ctx=36190, majf=0, minf=1016

  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 rwt: total=0,18050,0, short=0,0,0, dropped=0,0,0

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

 

Run status group 0 (all jobs):

  WRITE: bw=1200KiB/s (1229kB/s), 1200KiB/s-1200KiB/s (1229kB/s-1229kB/s), io=70.5MiB (73.9MB), run=60168-60168msec

 

Disk stats (read/write):

  rbd0: ios=0/18050, merge=0/11, ticks=0/59434, in_queue=59422, util=98.68%

_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux