lkml.org 
[lkml]   [2001]   [Oct]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: VM: 2.4.10 vs. 2.4.10-ac2 and qsort()
Hi guys,

I've already expressed my concern about using srand(1) in private e-mails.
I think it's unscientific to use one particular random sequence. Since
no one checked if that matters, I changed srand(1) to srand(time(NULL))
and I'm posting my results. I don't do testing of Alan or Linus's kernels,
but use recent Red Hat kernel. I think I've shown that it does matter.

Six quick consecutive runs of modified qs on a small set of 8 million
integers (obviously no swap activity):

> time ./a.out 8000000
0 errors.
24.250u 0.310s 0:24.55 100.0% 0+0k 0+0io 116pf+0w
0 errors.
24.290u 0.260s 0:24.55 100.0% 0+0k 0+0io 116pf+0w
0 errors.
24.300u 0.260s 0:24.55 100.0% 0+0k 0+0io 116pf+0w
0 errors.
24.270u 0.300s 0:24.57 100.0% 0+0k 0+0io 116pf+0w
0 errors.
24.290u 0.270s 0:24.56 100.0% 0+0k 0+0io 116pf+0w
0 errors.
24.280u 0.280s 0:24.55 100.0% 0+0k 0+0io 116pf+0w

Apparently, no significant deviations in computing times.

Six runs of modified qs on a large set of 80 million integers (a lot of
swapping!)

> time ./a.out 80000000
0 errors.
261.580u 4.250s 11:09.21 39.7% 0+0k 0+0io 17379pf+0w
0 errors.
260.460u 3.660s 9:09.72 48.0% 0+0k 0+0io 13194pf+0w
0 errors.
260.620u 4.510s 10:39.80 41.4% 0+0k 0+0io 16714pf+0w
0 errors.
261.790u 4.150s 10:09.58 43.6% 0+0k 0+0io 16331pf+0w
0 errors.
260.400u 4.140s 9:23.46 46.9% 0+0k 0+0io 13722pf+0w
0 errors.
259.980u 3.940s 9:10.22 47.9% 0+0k 0+0io 14240pf+0w

mean = 9m57s; standard deviation = 50s.

Apparently, the random sequence does matter (to the Rik's algorithm at
least since it's in RH kernel).

I wonder how big the deviation is for official and AC trees.
Now Lorenzo's results seem inconclusive.

Regards,
Alexei

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

\
 
 \ /
  Last update: 2005-03-22 13:04    [W:0.216 / U:0.396 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site