On Wed, 9 Feb 2000, Andrea Arcangeli wrote:
> Ok here it is:
>
> ftp://ftp.*.kernel.org/pub/linux/kernel/people/andrea/patches/
v2.2/2.2.14/elevator-starvation-4.gz
> ftp://ftp.*.kernel.org/pub/linux/kernel/people/andrea/patches/
v2.3/2.3.42/elevator-starvation-4.gz
>
> Please Lawrence drop the two preliminary versions I sent you by email
> privately, these new one are better (and also compiles :>). I spent some
> more time doing better tuning too.
Luckily I was feeling lazy that day. Yeah, it's always nice to have
things that actually compile. ;)
> I am interested if you could do a new bench of the rewrite case, I did
> something exactly to try to address the little drop of performance in the
> rewrite case.
Heres the new figures. My poor harddisk must have aged a couple of years
for all the testing I've done lately.
-------Sequential Output-------- ---Sequential Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
2.2.14
500 5294 87.2 10799 9.3 4305 8.4 5991 90.0 10568 7.4 93.5 1.6
2.2.14-aa2
500 6047 99.1 10840 9.4 3267 6.2 5381 80.7 11606 7.9 63.9 1.4
2.2.14-aa4
500 6045 99.1 11441 9.6 3022 6.2 4243 64.1 11528 8.3 89.8 1.6
2.2.14+elevator-2
500 6014 98.9 11460 9.9 3414 7.4 5313 80.1 11454 8.4 95.8 2.0
2.2.14+elevator-4 *NEW*
500 5988 98.6 10970 9.3 3913 7.8 5977 90.0 11150 8.1 92.4 1.6
2.3.42
500 6061 98.8 11995 9.2 4291 5.5 5527 82.7 11322 8.3 89.0 1.5
2.4.42+elevator-4 *NEW*
500 6065 98.9 10892 8.5 2556 3.5 5012 75.1 11046 8.7 86.5 1.4
Seems 2.2.14 elevator 4 has rewrite cured! Mostly. Dont know what
happened on the 2.3 equivilant run though? I will rerun the test tomorrow
and if I get different figures (I shouldnt, bonnie seems fairly consistant
on those 500meg runs.. certainly no more then +/-3%) I will let you know.
Machine feels smoothish with the big dd going on. I had a big dd along
with 4 compiles going and load up near 8 but interactive response was
still easily tollerable!!! Happy happy... All on one disk of course. I
also had a mp3 playing (xmms) and it didn't skip once.
Lawrence
-
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 : Tue Feb 15 2000 - 21:00:16 EST