Re: postgresql 8.3 tps rate - Mailing list pgsql-performance

From Ibrahim Harrani
Subject Re: postgresql 8.3 tps rate
Date
Msg-id 530068a0901222152y425a7455se68114b0ac4331ce@mail.gmail.com
Whole thread Raw
In response to Re: postgresql 8.3 tps rate  (Craig James <craig_james@emolecules.com>)
Responses Re: postgresql 8.3 tps rate
Re: postgresql 8.3 tps rate
Re: postgresql 8.3 tps rate
List pgsql-performance
Hi Craig,

Here is the result. It seems that disk write is terrible!.

root@myserver /usr]#  time (dd if=/dev/zero of=bigfile bs=8192
count=1000000; sync)


1000000+0 records in
1000000+0 records out
8192000000 bytes transferred in 945.343806 secs (8665630 bytes/sec)

real    15m46.206s
user    0m0.368s
sys     0m15.560s
[root@myserver /usr]#

[root@myserver /usr]#  time dd if=bigfile of=/dev/null bs=8192
1000000+0 records in
1000000+0 records out
8192000000 bytes transferred in 174.646798 secs (46906099 bytes/sec)

real    2m54.663s
user    0m0.246s
sys     0m9.307s


On Fri, Jan 23, 2009 at 12:44 AM, Craig James
<craig_james@emolecules.com> wrote:
> David Rees wrote:
>>
>> On Thu, Jan 22, 2009 at 1:27 PM, Ibrahim Harrani
>> <ibrahim.harrani@gmail.com> wrote:
>>>
>>> Version 1.93d       ------Sequential Output------ --Sequential Input-
>>> --Random-
>>> Concurrency   1     -Per Chr- --Block-- -Rewrite- -Per Chr- --Block--
>>> --Seeks--
>>> Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP
>>>  /sec %CP
>>> myserver 300M   391  97  9619   1  8537   2   673  99 +++++ +++  1196  16
>>> Latency               211ms     388ms     325ms   27652us     722us
>>>  6720ms
>>> Version 1.93d       ------Sequential Create------ --------Random
>>> Create--------
>>> myserver    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
>>>             files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec
>>> %CP
>>>                16  9004  25 +++++ +++ +++++ +++  8246  20 +++++ +++ +++++
>>> +++
>>> Latency               592ms     208us     102us     673ms     179us
>>> 100us
>>
>> You should be testing bonnie with a file size that is at least double
>> the amount of memory in your machine - in this case, 4GB files, not
>> 300MB files.
>>
>>> When I compare my  bonnie++ result with the one at
>>> http://www.westnet.com/~gsmith/content/postgresql/pg-disktesting.htm,
>>> It seems that there is something wrong with the disks!?
>>
>> Yes, your machine appears to be very slow.  You should be able to
>> write in the order of 30-50MB/s+ and read in the order of 40-80MB/s+.
>> Random IO should be in the 200 tps range for a 7200rpm SATA RAID1.
>
> Have you tried the really basic speed test?
>
>  time (dd if=/dev/zero of=bigfile bs=8192 count=1000000; sync)
>
>  time dd if=bigfile of=/dev/null bs=8192
>
> Divide 8.2GB by the times reported.  On a single 10K SATA drive, I get about
> 55MB/sec write and 61 MB/sec read.
>
> If you can't get similar numbers, then something is wrong.
>
> Craig
>

pgsql-performance by date:

Previous
From: Craig Ringer
Date:
Subject: Re: caching indexes and pages?
Next
From: Mark Wong
Date:
Subject: Re: dbt-2 tuning results with postgresql-8.3.5