Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 09 Aug 2007 16:39:18 +1000
From:      Michael Pope <michael@dtcorp.com.au>
To:        freebsd-performance@freebsd.org
Subject:   Bonnie++ test WD Raptor seems slow
Message-ID:  <46BAB696.5000204@dtcorp.com.au>

next in thread | raw e-mail | index | archive | help
I've just went a purchased a brand new WD Raptor 74GB WD740ADFD and 
compared my bonnie++ test I did against a Samsung 120GB 7200rpm drive 
and the Raptor seems slow.

Here is the command I used to test
$ bonnie++ -s 1000

Here are the results from the Raptor
Version 1.93c       ------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
paistram2.lan 1000M   266  99 43993  17  6827   2   481  99 27622   7  
51.4   2
Latency             57509us    1458ms    2304ms   33107us     595ms    
7295ms
Version 1.93c       ------Sequential Create------ --------Random 
Create--------
paistram2.lan       -Create-- --Read--- -Delete-- -Create-- --Read--- 
-Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  
/sec %CP
                 16  1275   5 +++++ +++ 31905  99  4855  22 32455  94 
31986  99
Latency              1786ms    2264us      82us    1004ms   24603us      
89us
1.93c,1.93c,paistram2.lan,1,1186637120,1000M,,266,99,43993,17,6827,2,481,99,27622,7,51.4,2,16,,,,,1275,5,+++++,+++,31905,99,4855,22,32455,94,31986,99,57509us,1458ms,2304ms,33107us,595ms,7295ms,1786ms,2264us,82us,1004ms,24603us,89us


Here are the results for a Samsung 120GB 7200rpm drive:
Version 1.93c       ------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
ws10.pclan    1000M   343  86 53058  12 19133   4   721  88 51198   7 
129.3   3
Latency             66491us     190ms     490ms   53621us   28599us    
2679ms
Version 1.93c       ------Sequential Create------ --------Random 
Create--------
ws10.pclan          -Create-- --Read--- -Delete-- -Create-- --Read--- 
-Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  
/sec %CP
                 16 13805  26 +++++ +++ +++++ +++ 23221  47 +++++ +++ 
+++++ +++
Latency               687ms    2506us    2616us     101ms    2543us    
2742us
1.93c,1.93c,ws10.pclan,1,1186620180,1000M,,343,86,53058,12,19133,4,721,88,51198,7,129.3,3,16,,,,,13805,26,+++++,+++,+++++,+++,23221,47,+++++,+++,+++++,+++,66491us,190ms,490ms,53621us,28599us,2679ms,687ms,2506us,2616us,101ms,2543us,2742us



I'm not sure if I'm reading this right, but the Samsung drive looks like 
it's killing the Raptor in Sequential Output and Input.
Has anyone got any ideas?

I'm just using the standard onboard controller (Intel ICH6), but I will 
be using a 3ware 9500S-4LP in the end.


-- 
Michael



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?46BAB696.5000204>