Re: IOZone Performance is very Strange

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

 



Hi,

We choose EXT4 for stability instead of btrfs.
If we try btrfs later, I'll provide the throughput for you. :)

In fact,  I still think the iozone result is unusual, is anyone has
idea with the read/write performance? (write: 1~2 GB/s)

Thanks a lot~

2011/5/17 Simon Tian <aixt2006@xxxxxxxxx>:
> Hi,
>        The throughput of EXT4 is nice.... How about btrfs?
>
> Thx!
> Simon
>
> 2011/5/17 AnnyRen <annyren6@xxxxxxxxx>:
>> Hi, Jeff:
>>
>> I run "ceph osd tell osd_num bench" with 1 times per osd
>>
>> and use ceph -w to observe every osd performance,
>>
>> osd0:  [INF] bench: wrote 1024 MB in blocks of 4096 KB in 10.875844
>> sec at 96413 KB/sec
>> osd1:  [INF] bench: wrote 1024 MB in blocks of 4096 KB in 11.784985
>> sec at 88975 KB/sec
>> osd2:  [INF] bench: wrote 1024 MB in blocks of 4096 KB in 11.161067
>> sec at 93949 KB/sec
>> osd3:  [INF] bench: wrote 1024 MB in blocks of 4096 KB in 10.798796
>> sec at 97101 KB/sec
>> osd4:  [INF] bench: wrote 1024 MB in blocks of 4096 KB in 14.437141
>> sec at 72630 KB/sec
>> osd5:  [INF] bench: wrote 1024 MB in blocks of 4096 KB in 14.451444
>> sec at 72558 KB/sec
>> osd6:  [INF] bench: wrote 1024 MB in blocks of 4096 KB in 11.083872
>> sec at 94603 KB/sec
>> osd7:  [INF] bench: wrote 1024 MB in blocks of 4096 KB in 11.062728
>> sec at 94784 KB/sec
>> osd8:  [INF] bench: wrote 1024 MB in blocks of 4096 KB in 14.137312
>> sec at 74170 KB/sec
>> osd9:  [INF] bench: wrote 1024 MB in blocks of 4096 KB in 13.489992
>> sec at 77729 KB/sec
>>
>>
>> and I run
>> root@MDS2:/mnt/ceph# rados bench 60 write -p data
>>
>> the result is
>>
>> Total time run:        60.553247
>> Total writes made:     1689
>> Write size:            4194304
>> Bandwidth (MB/sec):    111.571
>>
>> Average Latency:       0.573209
>> Max latency:           2.25691
>> Min latency:           0.218494
>>
>>
>>
>> 2011/5/17 Jeff Wu <cpwu@xxxxxxxxxxxxx>:
>>> Hi AnnyRen
>>>
>>> Could you run the following commands and give us the test results?
>>>
>>> $ceph osd tell OSD-N bench    // OSD-N : osd number : 0,1,2 ....
>>> $ceph -w
>>>
>>> $rados bench 60 write -p data    // refer to "rados -h "
>>>
>>> Jeff
>>>
>>>
>>>
>>> On Tue, 2011-05-17 at 11:53 +0800, AnnyRen wrote:
>>>> I'm running iozone on EXT4 with Ceph v0.26.
>>>> But I got the weird result, most write performance exceed 1GB/s, even
>>>> up to 3GB/s
>>>> I think it's not normal to get the performance outpupt.
>>>>
>>>> Command line I used is: /usr/bin/iozone -azcR -f
>>>> /mnt/ceph/test0516.dbf -g 1G -b /exceloutput0516.xls
>>>> Attachment is the output file...
>>>>
>>>> and my environment is composed of 15 physical machines with
>>>>
>>>> 3 MON, 2MDS (1 active, 1 standby), 10 OSD (1 osd daemon (3T) /host)
>>>> EXT4 format
>>>> data replication size: 3
>>>>
>>>>
>>>>
>>>> "Writer report"
>>>>         "4"  "8"  "16"  "32"  "64"  "128"  "256"  "512"  "1024"
>>>> "2048"  "4096"  "8192"  "16384"
>>>> "64"   983980  1204796  1210227  1143223  1357066
>>>> "128"   1007629  1269781  1406136  1391557  1436229  1521718
>>>> "256"   1112909  1430119  1523457  1652404  1514860  1639786  1729594
>>>> "512"   1150351  1475116  1605228  1723770  1797349  1712772  1783912  1854787
>>>> "1024"   1213334  1471481  1679160  1828574  1888889  1899750  1885572
>>>>  1865912  1875690
>>>> "2048"   1229274  1540849  1708146  1843410  1903457  1980705  1930406
>>>>  1913634  1906837  1815744
>>>> "4096"   1213284  1528348  1674646  1762434  1872096  1882352  1881528
>>>>  1903416  1897949  1835102  1731177
>>>> "8192"   204560  155186  572387  238548  186597  429036  187327
>>>> 157205  553771  416512  299810  405842
>>>> "16384"   699749  559255  687450  541030  828776  555296  742561
>>>> 525483  604910  452423  564557  670539  970616
>>>> "32768"   532414  829610  812215  879441  863159  864794  865938
>>>> 804951  916352  879582  608132  860732  1239475
>>>> "65536"   994824  1096543  1095791  1317968  1280277  1390267  1259868
>>>>  1205214  1339111  1346927  1267888  863234  1190221
>>>> "131072"   1063429  1165115  1102650  1554828  1182128  1185731
>>>> 1190752  1195792  1277441  1211063  1237567  1226999  1336961
>>>> "262144"   1280619  1368554  1497837  1633397  1598255  1609212
>>>> 1607504  1665019  1590515  1548307  1591258  1505267  1625679
>>>> "524288"   1519583  1767928  1738523  1883151  2011216  1993877
>>>> 2023543  1867440  2106124  2055064  1906668  1778645  1838988
>>>> "1048576"   1580851  1887530  2044131  2166133  2236379  2283578
>>>> 2257454  2296612  2271066  2101274  1905829  1605923  2158238
>>>>
>>>>
>>>>
>>>> "Reader report"
>>>>         "4"  "8"  "16"  "32"  "64"  "128"  "256"  "512"  "1024"
>>>> "2048"  "4096"  "8192"  "16384"
>>>> "64"   1933893  2801873  3057153  3363612  3958892
>>>> "128"   2286447  3053774  2727923  3468030  4104338  4557257
>>>> "256"   2903529  3236056  3245838  3705040  3654598  4496299  5117791
>>>> "512"   2906696  3437042  3628697  3431550  4871723  4296637  6246213  6395018
>>>> "1024"   3229770  3483896  4609294  3791442  4614246  5536137  4550690
>>>>  5048117  4966395
>>>> "2048"   3554251  4310472  3885431  4096676  6401772  4842658  5080379
>>>>  5184636  5596757  5735012
>>>> "4096"   3416292  4691638  4321103  5728903  5475122  5171846  4819300
>>>>  5258919  6408472  5044289  4079948
>>>> "8192"   3233004  4615263  4536055  5618186  5414558  5025700  5553712
>>>>  4926264  5634770  5281396  4659702  3652258
>>>> "16384"   3141058  3704193  4567654  4395850  4568869  5387732
>>>> 4436432  5808029  5578420  4675810  3913007  3911225  3961277
>>>> "32768"   3704273  4598957  4088278  5133719  5896692  5537024
>>>> 5234412  5398271  4942992  4118662  3729099  3511757  3481511
>>>> "65536"   4131091  4210184  5341188  4647619  6077765  5852474
>>>> 5379762  5259330  5488249  5246682  4342682  3549202  3286487
>>>> "131072"   3582876  5251082  5332216  5269908  5303512  5574440
>>>> 5635064  5796372  5406363  4958839  4435918  3673443  3647874
>>>> "262144"   3659283  4551414  5746231  5433824  5876196  6011650
>>>> 5552000  5629260  5298830  4982226  4628902  4065823  3421924
>>>> "524288"   3905973  5488778  5219963  6047356  5916811  6180455
>>>> 5495733  5925628  5637512  5537123  3517132  3550861  3047013
>>>> "1048576"   3855595  5634571  5410298  6001809  6464713  6299610
>>>> 5894249  5516031  5800161  5209203  4295840  3724983  3641623
>>>
>>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
>> the body of a message to majordomo@xxxxxxxxxxxxxxx
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
>

Attachment: outputlog0516.log
Description: Binary data


[Index of Archives]     [CEPH Users]     [Ceph Large]     [Information on CEPH]     [Linux BTRFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]
  Powered by Linux