Re: I/O performance on 2.4.0-test2

From: Jens Axboe (axboe@suse.de)
Date: Sun Jun 25 2000 - 07:59:50 EST

On Sun, Jun 25 2000, Derek Martin wrote:
> I haven't been watching it too closely, but I just did some bonnie
> comparisons on my 2.2.15 kernel, and my 2.4.0-test2 kernel, and here's
> what I got for results:

I haven't used bonnie to do testing, so I tried bonnie++ and compared
2.4.0-test2 to 2.2.17-pre16. At first the results were a bit
discouraging (when the page cache filled, performance dropped
drastically), so I applied ac1 which has a lot of the vm tweaks from
the ac series.

2.2.17-pre6

Version 0.99j ------Sequential Output------ --Sequential Input- --Random-
               -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine MB K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
Unknown 1024 6563 96 14945 20 6585 16 7009 95 15382 10 12.1 0
               ------Sequential Create------ --------Random Create--------
               -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
         files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
            30 145 99 477 99 4684 97 146 99 618 99 491 78

2.4.0-test2-ac1

Version 0.99j ------Sequential Output------ --Sequential Input- --Random-
               -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine MB K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
Unknown 1024 6564 96 17327 22 6840 9 6119 86 15479 11 12.1 0
               ------Sequential Create------ --------Random Create--------
               -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
         files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
            30 129 99 +++++ 103 5075 99 131 99 +++++ 98 774 92

Then it looks pretty decent, but there's still a bit of vm work to
be done. Note the ++++ results, apparently this means that the test
finished in < 1 second, so relevant numbers cannot be printed.

I also did a dbench run.

2.2.17-pre6

burns:/opt/software/testing # ./dbench 48
48 clients started
Throughput 12.84 MB/sec (NB=16.0499 MB/sec 128.4 MBit/sec)

2.4.0-test2-ac1

burns:/opt/software/testing # ./dbench 48
48 clients started
Throughput 30.6317 MB/sec (NB=38.2896 MB/sec 306.317 MBit/sec)

The 2.2 results are very disappointing here, the last time I checked
2.2 dbench results were in the ~24MB/sec range. 

-- 
* Jens Axboe <axboe@suse.de>
* SuSE Labs

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

This archive was generated by hypermail 2b29 : Mon Jun 26 2000 - 21:00:06 EST