Re: raid5 performance after upgrade to 2.4.28

[Tim Moore <linux-raid@nsr500.net>]

Did values change in /proc/sys/vm/{min,max}-readahead?

http://lists.us.dell.com/pipermail/linux-poweredge/2004-February/036166.html


Guy wrote:
> Now the bad news...
> 
> I have 2 CPUs  P3-500Mhz, 512Meg ram.  So I ran 2 bonnies at the same time.
> 
> Kernel 2.4.20-8 with 2 Bonnies at the same time:
> Version  1.03 ------Sequential Output------ --Sequential Input- --Random-
>               -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
> watkins-ho 1G  3091  85 18593  44  9733  24  3443  97 59895  60 249.6   6
> watkins-ho 1G  2980  87 21176  54 10167  23  3478  99 44525  44 384.2   9
>                ----     -----     -----      ----    ------     -----
> Total          6071     39769     19900      6921    104420     633.8
> 
> Kernel 2.4.28 with 2 Bonnies at the same time:
> watkins-ho 1G  4456  92 17922  52  6489  17  4961  96 26571  32 320.4  12
> watkins-ho 1G  4343  89 17987  52  6197  17  4946  96 25646  31 363.7   7
>                ----     -----     -----      ----    ------     -----
> Total          8799     35909     12686      9907     52217     684.1
> 
> Very odd, the sequential read speed is almost exactly 1/2 now.
> 
> Oh well, I still can't tell without a tool to measure the speed.
> But I still don't like it!
> 
> My network is only 100BaseT, so it is still well beyond that!
> 
> Guy
> 
> -----Original Message-----
> From: linux-raid-owner@vger.kernel.org
> [mailto:linux-raid-owner@vger.kernel.org] On Behalf Of Guy
> Sent: Wednesday, January 12, 2005 1:17 AM
> To: linux-raid@vger.kernel.org
> Subject: raid5 performance after upgrade to 2.4.28
> 
> About 1 month ago I upgraded from Kernel 2.4.20-8 to 2.4.28.
> Today I noticed my sequential read speed is up by about 12%.
> 
> I don't recall making any other changes that would affect speed.
> So, I guess the Kernel upgrade made the difference.
> 
> An old bonnie output:
> Version  1.03 ------Sequential Output------ --Sequential Input- --Random-
>               -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
> watkins-ho 1G  3293  97 34587  89 22306  53  3549  99 67492  63 500.1   9
> 
> Today:
> watkins-ho 1G  4944  99 31719  65 18538  43  5184  98 75440  71 602.5   8
>                ^^^^                          ^^^^     ^^^^^     ^^^^^
> 
> Some things improved, others decreased.
> 
> Of course, no change I can notice on my own.
> I needed to measure performance to notice.
> 
> I guess I should try Kernel 2.6 some day.
> 
> Guy
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-raid" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-raid" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 


-- 
  | for direct mail add "private_" in front of user name