Hi All,
I tried bumping my read ahead up to 4096. Instead of having faster reads, it
seems it actually slowed things down. In fact, most of the tuning suggestions
I've tried have made little to no difference in the results I get from
bonnie++. I'll include a table of values in html. I'm wondering if these are
normal values in my case; 4 disk RAID10 Linux ext3 146GB SAS 15K RPM Drive.
Title: Benchmarks
4 disk RAID 10 ext3 Red Hat
| Config |
|
Block Out |
Rewrite Out |
Block In |
Random Seeks |
| Read Ahead |
noatime |
swappiness |
overcommit |
dirty ratio |
dirty bg ratio |
Chunk size |
K/sec |
%cpu |
K/sec |
%cpu |
K/sec |
%cpu |
/sec |
%cpu |
| 192 |
noatime |
0 |
2 |
2 |
1 |
96280M |
163189 |
29 |
102625 |
17 |
288505 |
17 |
583.4 |
1 |
| 256 |
default |
default |
default |
40 |
10 |
96280M |
160848 |
28 |
105445 |
17 |
288802 |
17 |
588.2 |
1 |
| 256 |
noatime |
default |
default |
40 |
10 |
96280M |
157318 |
26 |
106947 |
17 |
289369 |
17 |
603.7 |
1 |
| 256 |
noatime |
0 |
2 |
2 |
1 |
96280M |
162137 |
29 |
104220 |
17 |
292700 |
17 |
609.3 |
1 |
| 1024 |
noatime |
0 |
2 |
2 |
1 |
96280M |
162908 |
29 |
98724 |
15 |
267720 |
13 |
665.0 |
0 |
| 4096 |
default |
default |
default |
40 |
10 |
96280M |
160507 |
28 |
105678 |
15 |
277123 |
13 |
666.5 |
1 |
| 4096 |
noatime |
default |
default |
40 |
10 |
96280M |
159806 |
27 |
106396 |
15 |
276359 |
13 |
525.1 |
1 |
| 8192 |
noatime |
default |
default |
40 |
10 |
96280M |
122381 |
21 |
103858 |
15 |
270377 |
13 |
658.0 |
1 |
Thank you,
Ramsey
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
