| 1 |
On Fri, Jul 19, 2013 at 08:34:31PM -0300, luis jure wrote: |
| 2 |
> on 2013-07-19 at 17:11 Bruce Hill wrote: |
| 3 |
> |
| 4 |
> > Do some meaningful benchmark... |
| 5 |
> > |
| 6 |
> > emerge -ajv app-benchmarks/bonnie++ && bonnie++ -d /tmp -u root |
| 7 |
> > |
| 8 |
> > then post us your output. |
| 9 |
> |
| 10 |
> mmm... /tmp is on the root partition that's only about 40 Gb (in my old |
| 11 |
> HDD / was 30Gb, and that was enough, but sometimes i had to clean a bit to |
| 12 |
> compile libreoffice, that checks for 6Gb free). |
| 13 |
> |
| 14 |
> apparently that's not enough for bonnie++ to run, so i mounted the other |
| 15 |
> partition in the SSD and performed the benchmark there. |
| 16 |
> |
| 17 |
> here's the result : |
| 18 |
> |
| 19 |
> |
| 20 |
> Using uid:0, gid:0. |
| 21 |
> Writing a byte at a time...done |
| 22 |
> Writing intelligently...done |
| 23 |
> Rewriting...done |
| 24 |
> Reading a byte at a time...done |
| 25 |
> Reading intelligently...done |
| 26 |
> start 'em...done...done...done...done...done... |
| 27 |
> Create files in sequential order...done. |
| 28 |
> Stat files in sequential order...done. |
| 29 |
> Delete files in sequential order...done. |
| 30 |
> Create files in random order...done. |
| 31 |
> Stat files in random order...done. |
| 32 |
> Delete files in random order...done. |
| 33 |
> Version 1.97 ------Sequential Output------ --Sequential Input- |
| 34 |
> --Random- Concurrency 1 -Per Chr- --Block-- -Rewrite- -Per Chr- |
| 35 |
> --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP |
| 36 |
> K/sec %CP K/sec %CP /sec %CP acme7 24016M 571 98 261698 22 |
| 37 |
> 133029 14 4272 99 287927 12 8057 53 Latency 23779us |
| 38 |
> 78780us 591ms 2378us 51825us 892ms Version 1.97 |
| 39 |
> ------Sequential Create------ --------Random Create-------- |
| 40 |
> acme7 -Create-- --Read--- -Delete-- -Create-- --Read--- |
| 41 |
> -Delete-- files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec |
| 42 |
> %CP 16 18593 19 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++ |
| 43 |
> Latency 2033us 419us 1322us 109us 4us |
| 44 |
> 33us |
| 45 |
> 1.97,1.97,acme7,1,1374260412,24016M,,571,98,261698,22,133029,14,4272,99,287927,12,8057,53,16,,,,,18593,19,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,23779us,78780us,591ms,2378us,51825us,892ms,2033us,419us,1322us,109us,4us,33us |
| 46 |
> |
| 47 |
> |
| 48 |
> can you read that? i'm afraid i don't know much what those numbers mean in |
| 49 |
> terms of performance... |
| 50 |
|
| 51 |
Here's the definitive guide on how to understand it: |
| 52 |
|
| 53 |
https://blogs.oracle.com/roch/entry/decoding_bonnie |
| 54 |
|
| 55 |
Anyway, now you've got something to compare different disks on your hardware. |
| 56 |
|
| 57 |
Cheers, |
| 58 |
Bruce |
| 59 |
-- |
| 60 |
Happy Penguin Computers >') |
| 61 |
126 Fenco Drive ( \ |
| 62 |
Tupelo, MS 38801 ^^ |
| 63 |
support@×××××××××××××××××××××.com |
| 64 |
662-269-2706 662-205-6424 |
| 65 |
http://happypenguincomputers.com/ |
| 66 |
|
| 67 |
A: Because it messes up the order in which people normally read text. |
| 68 |
Q: Why is top-posting such a bad thing? |
| 69 |
A: Top-posting. |
| 70 |
Q: What is the most annoying thing in e-mail? |
| 71 |
|
| 72 |
Don't top-post: http://en.wikipedia.org/wiki/Top_post#Top-posting |
Archives