1 |
Hey guys! |
2 |
|
3 |
Yesterday I found this article |
4 |
http://gentoo-wiki.com/HOWTO_Speed_up_decompression_with_pbzip2 and was |
5 |
very happy, because I'm an owner of an Q6600. |
6 |
|
7 |
So I tried to reproduce the benchmark from the wiki article but got that |
8 |
bad results: |
9 |
|
10 |
test # time *bzip2* -d -kf linux-2.6.23.tar.bz2 |
11 |
|
12 |
real 0m11.672s |
13 |
user 0m11.306s |
14 |
sys 0m0.367s |
15 |
|
16 |
|
17 |
test # time *pbzip2* -d -p4 -kf linux-2.6.23.tar.bz2 |
18 |
|
19 |
real 0m25.554s |
20 |
user 0m24.862s |
21 |
sys 0m0.683s |
22 |
|
23 |
So the parallel version took more than the double time! |
24 |
|
25 |
To test whether this is a problem of my Pc I tested this on an Dual Core |
26 |
with the same result. |
27 |
|
28 |
An test with 7z was much better: |
29 |
|
30 |
|
31 |
test # time *7za* x -y linux-2.6.23.tar.bz2 |
32 |
|
33 |
real 0m4.642s |
34 |
user 0m8.379s |
35 |
sys 0m0.327s |
36 |
|
37 |
|
38 |
All tests where done in a tmpfs off 1GB on 2GB RAM. |
39 |
|
40 |
|
41 |
So my questions is what did I do wrong? |
42 |
|
43 |
I also tested it with different CFLAGS, but my CFLAGS are basically very |
44 |
conservative. (-O2 -march=prescott -pipe -fomit-frame-pointer) |