pg.ddx.io pgsql-performance@postgresql.org mailing list archive
help / color / mirror / Atom feedSoftware vs. Hardware RAID Data
5+ messages / 3 participants
[nested] [flat]
* Software vs. Hardware RAID Data
@ 2008-08-20 05:23 Mark Wong <markwkm@gmail.com>
0 siblings, 2 replies; 5+ messages in thread
From: Mark Wong @ 2008-08-20 05:23 UTC (permalink / raw)
To: pgsql-performance; +Cc: Gabrielle Roth <gorthx@gmail.com>
Hi all,
We started an attempt to slice the data we've been collecting in
another way, to show the results of software vs. hardware RAID:
http://wiki.postgresql.org/wiki/HP_ProLiant_DL380_G5_Tuning_Guide#Hardware_vs._Software_Raid
The angle we're trying to show here is the processor utilization and
i/o throughput for a given file system and raid configuration. I
wasn't sure about the best way to present it, so this is how it looks
so far. Click on the results for a chart of the aggregate processor
utilization for the test.
Comments, suggestions, criticisms, et al. welcome.
Regards,
Mark
^ permalink raw reply [nested|flat] 5+ messages in thread
* Re: Software vs. Hardware RAID Data
@ 2008-08-20 05:49 david@lang.hm
parent: Mark Wong <markwkm@gmail.com>
1 sibling, 1 reply; 5+ messages in thread
From: david@lang.hm @ 2008-08-20 05:49 UTC (permalink / raw)
To: Mark Wong <markwkm@gmail.com>; +Cc: pgsql-performance; Gabrielle Roth <gorthx@gmail.com>
On Tue, 19 Aug 2008, Mark Wong wrote:
> Hi all,
>
> We started an attempt to slice the data we've been collecting in
> another way, to show the results of software vs. hardware RAID:
>
> http://wiki.postgresql.org/wiki/HP_ProLiant_DL380_G5_Tuning_Guide#Hardware_vs._Software_Raid
>
> The angle we're trying to show here is the processor utilization and
> i/o throughput for a given file system and raid configuration. I
> wasn't sure about the best way to present it, so this is how it looks
> so far. Click on the results for a chart of the aggregate processor
> utilization for the test.
>
> Comments, suggestions, criticisms, et al. welcome.
it's really good to show cpu utilization as well as throughput, but how
about showing the cpu utilization as %cpu per MB/s (possibly with a flag
to indicate any entries that look like they may have hit cpu limits)
why did you use 4M stripe size on the software raid? especially on raid 5
this seems like a lot of data to have to touch when making an update.
David Lang
^ permalink raw reply [nested|flat] 5+ messages in thread
* Re: Software vs. Hardware RAID Data
@ 2008-08-20 07:53 Tommy Gildseth <tommy.gildseth@usit.uio.no>
parent: Mark Wong <markwkm@gmail.com>
1 sibling, 1 reply; 5+ messages in thread
From: Tommy Gildseth @ 2008-08-20 07:53 UTC (permalink / raw)
To: Mark Wong <markwkm@gmail.com>; +Cc: pgsql-performance
Mark Wong wrote:
> Hi all,
>
> We started an attempt to slice the data we've been collecting in
> another way, to show the results of software vs. hardware RAID:
>
> http://wiki.postgresql.org/wiki/HP_ProLiant_DL380_G5_Tuning_Guide#Hardware_vs._Software_Raid
>
> Comments, suggestions, criticisms, et al. welcome.
The link to the graph for "Two Disk Software RAID-0 (64KB stripe)"
points to the wrong graph, hraid vs sraid.
--
Tommy Gildseth
^ permalink raw reply [nested|flat] 5+ messages in thread
* Re: Software vs. Hardware RAID Data
@ 2008-08-20 15:02 Mark Wong <markwkm@gmail.com>
parent: david@lang.hm
0 siblings, 0 replies; 5+ messages in thread
From: Mark Wong @ 2008-08-20 15:02 UTC (permalink / raw)
To: david@lang.hm; +Cc: pgsql-performance; Gabrielle Roth <gorthx@gmail.com>
On Tue, Aug 19, 2008 at 10:49 PM, <david@lang.hm> wrote:
> On Tue, 19 Aug 2008, Mark Wong wrote:
>
>> Hi all,
>>
>> We started an attempt to slice the data we've been collecting in
>> another way, to show the results of software vs. hardware RAID:
>>
>>
>> http://wiki.postgresql.org/wiki/HP_ProLiant_DL380_G5_Tuning_Guide#Hardware_vs._Software_Raid
>>
>> The angle we're trying to show here is the processor utilization and
>> i/o throughput for a given file system and raid configuration. I
>> wasn't sure about the best way to present it, so this is how it looks
>> so far. Click on the results for a chart of the aggregate processor
>> utilization for the test.
>>
>> Comments, suggestions, criticisms, et al. welcome.
>
> it's really good to show cpu utilization as well as throughput, but how
> about showing the cpu utilization as %cpu per MB/s (possibly with a flag to
> indicate any entries that look like they may have hit cpu limits)
Ok, we'll add that and see how it looks.
> why did you use 4M stripe size on the software raid? especially on raid 5
> this seems like a lot of data to have to touch when making an update.
I'm sort of taking a shotgun approach, but ultimately we hope to show
whether there is significant impact of the stripe width relative to
the database blocksize.
Regards,
Mark
^ permalink raw reply [nested|flat] 5+ messages in thread
* Re: Software vs. Hardware RAID Data
@ 2008-08-20 15:05 Mark Wong <markwkm@gmail.com>
parent: Tommy Gildseth <tommy.gildseth@usit.uio.no>
0 siblings, 0 replies; 5+ messages in thread
From: Mark Wong @ 2008-08-20 15:05 UTC (permalink / raw)
To: Tommy Gildseth <tommy.gildseth@usit.uio.no>; +Cc: pgsql-performance
On Wed, Aug 20, 2008 at 12:53 AM, Tommy Gildseth
<tommy.gildseth@usit.uio.no> wrote:
> Mark Wong wrote:
>>
>> Hi all,
>>
>> We started an attempt to slice the data we've been collecting in
>> another way, to show the results of software vs. hardware RAID:
>>
>>
>> http://wiki.postgresql.org/wiki/HP_ProLiant_DL380_G5_Tuning_Guide#Hardware_vs._Software_Raid
>>
>> Comments, suggestions, criticisms, et al. welcome.
>
>
> The link to the graph for "Two Disk Software RAID-0 (64KB stripe)" points to
> the wrong graph, hraid vs sraid.
Thanks, I think I have it right this time.
Regards,
Mark
^ permalink raw reply [nested|flat] 5+ messages in thread
end of thread, other threads:[~2008-08-20 15:05 UTC | newest]
Thread overview: 5+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2008-08-20 05:23 Software vs. Hardware RAID Data Mark Wong <markwkm@gmail.com>
2008-08-20 05:49 ` david@lang.hm
2008-08-20 15:02 ` Mark Wong <markwkm@gmail.com>
2008-08-20 07:53 ` Tommy Gildseth <tommy.gildseth@usit.uio.no>
2008-08-20 15:05 ` Mark Wong <markwkm@gmail.com>
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox