[DRBD-user] Slow Read performance

Maxim Ianoglo dotnox at gmail.com
Mon May 2 20:33:10 CEST 2011

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


Hello,

With more testing and trying to use another I/O benchmark tool I am more
sure that the
bonnie++ is not a really good tool for benchmarking, it shows a little bit
strange results. Same I can say about iozone. So I did some test with "dd"
and reading from .dev.null to file, from file to .dev.null ( to emulate the
working of a web server for example and now the results are more
believable!. Reading with ~340-350MB/s. Writing: ~120-150MB/s

Thank you.
--
Maxim Ianoglo

On Mon, May 2, 2011 at 4:35 PM, Maxim Ianoglo <dotnox at gmail.com> wrote:

> No. This is output of DRBD state while performing test reading by bonnie++
> ( after test writin )
>
> There is no any "oos" data.
>
> [root at srv04 ~]$ cat /proc/drbd
> version: 8.3.10 (api:88/proto:86-96)
> GIT-hash: 5c0b0469666682443d4785d90a2c603378f9017b build by
> root at srv04.initalia.com, 2011-04-29 12:14:44
>  0: cs:Connected ro:Secondary/Primary ds:UpToDate/UpToDate C r-----
>     ns:0 nr:50388716 dw:50388716 dr:220 al:0 bm:2 lo:0 pe:0 ua:0 ap:0 ep:1
> wo:d oos:0
> [root at srv04 ~]$ cat /proc/drbd
> version: 8.3.10 (api:88/proto:86-96)
> GIT-hash: 5c0b0469666682443d4785d90a2c603378f9017b build by
> root at srv04.initalia.com, 2011-04-29 12:14:44
>  0: cs:Connected ro:Secondary/Primary ds:UpToDate/UpToDate C r-----
>     ns:0 nr:50388716 dw:50388716 dr:220 al:0 bm:2 lo:0 pe:0 ua:0 ap:0 ep:1
> wo:d oos:0
> [root at srv04 ~]$ cat /proc/drbd
> version: 8.3.10 (api:88/proto:86-96)
> GIT-hash: 5c0b0469666682443d4785d90a2c603378f9017b build by
> root at srv04.initalia.com, 2011-04-29 12:14:44
>  0: cs:Connected ro:Secondary/Primary ds:UpToDate/UpToDate C r-----
>     ns:0 nr:50388728 dw:50388728 dr:220 al:0 bm:2 lo:0 pe:0 ua:0 ap:0 ep:1
> wo:d oos:0
> [root at srv04 ~]$ cat /proc/drbd
> version: 8.3.10 (api:88/proto:86-96)
> GIT-hash: 5c0b0469666682443d4785d90a2c603378f9017b build by
> root at srv04.initalia.com, 2011-04-29 12:14:44
>  0: cs:Connected ro:Secondary/Primary ds:UpToDate/UpToDate C r-----
>     ns:0 nr:50388728 dw:50388728 dr:220 al:0 bm:2 lo:0 pe:0 ua:0 ap:0 ep:1
> wo:d oos:0
> [root at srv04 ~]$ cat /proc/drbd
> version: 8.3.10 (api:88/proto:86-96)
> GIT-hash: 5c0b0469666682443d4785d90a2c603378f9017b build by
> root at srv04.initalia.com, 2011-04-29 12:14:44
>  0: cs:Connected ro:Secondary/Primary ds:UpToDate/UpToDate C r-----
>     ns:0 nr:50388728 dw:50388728 dr:220 al:0 bm:2 lo:0 pe:0 ua:0 ap:0 ep:1
> wo:d oos:0
>
> --
> Maxim Ianoglo
>
>
>
> On Mon, May 2, 2011 at 2:59 PM, Bart Coninckx <bart.coninckx at telenet.be>wrote:
>
>> Because it is syncing?
>>
>>
>> On 05/02/11 13:45, Maxim Ianoglo wrote:
>>
>>> Hi,
>>>
>>> Anyway, why I get so big performance difference between ?
>>> On DRBD device with Secondary DRBD host connected: ~80MB/s Read and
>>> ~340MB/s Write
>>> On DRBD without Secondary DRBD host connected: ~245MB/s Read and
>>> ~340Write
>>>
>>> Thank you.
>>> --
>>> Maxim Ianoglo
>>>
>>>
>>>
>>> _______________________________________________
>>> drbd-user mailing list
>>> drbd-user at lists.linbit.com
>>> http://lists.linbit.com/mailman/listinfo/drbd-user
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20110502/b3d98800/attachment.htm>


More information about the drbd-user mailing list