View previous topic :: View next topic |
Author |
Message |
jsosic Guru
Joined: 02 Aug 2004 Posts: 510 Location: Split (Croatia)
|
Posted: Tue Aug 15, 2006 1:08 pm Post subject: Filesystem benchmark wierd results... |
|
|
I've begin a very intense FS performance test, and after testing only few filesystems, I've encountered really wierd results... I was planning to publish it in documentation and tips/tricks section of Gentoo Forums. My first test is writing speed test -> "time cp -a /usr /partition". /usr and /partition are on the different HDDs. Second test is reading speed test "time tar -cf - /partition | cat > /dev/null". One of the last tests is bonnie++, and for what it seems, bonnie++ gives me totally oposite results of the two previous tests. For example, copying 4.1gb /usr dir on stock XFS lasts 11m, and on JFS 14min, but JFS gets 5-10% better results on bonnie++ test. What do you think, why is this happening? _________________ I avenge with darkness, the blood is the life
The Order of the Dragon, I feed on human life |
|
Back to top |
|
|
sageman Guru
Joined: 04 May 2005 Posts: 363 Location: New Hampshire
|
Posted: Tue Aug 15, 2006 6:53 pm Post subject: |
|
|
How does bonnie++ determine the results? It is widely known that XFS is much more processor intensive than other filesystems, which might account for its lower score. _________________ Carlton Stedman
Gentoo Metalheads on Last.fm: http://www.last.fm/group/Gentoo+Metalheads |
|
Back to top |
|
|
Enlight Advocate
Joined: 28 Oct 2004 Posts: 3519 Location: Alsace (France)
|
Posted: Thu Aug 17, 2006 2:27 pm Post subject: |
|
|
sageman wrote: | How does bonnie++ determine the results? It is widely known that XFS is much more processor intensive than other filesystems, which might account for its lower score. |
@ kakashi senseï : Not that much, it depends of the number of AG's and log buffers. JFS is by far the less cpu intensive, But on today's machines hard drive is the bottleneck. Plus, I rather by far real world test in the spirit of what I could myself do, than random benchmark. |
|
Back to top |
|
|
sageman Guru
Joined: 04 May 2005 Posts: 363 Location: New Hampshire
|
Posted: Thu Aug 17, 2006 4:31 pm Post subject: |
|
|
Enlight wrote: | sageman wrote: | How does bonnie++ determine the results? It is widely known that XFS is much more processor intensive than other filesystems, which might account for its lower score. |
@ kakashi senseï : Not that much, it depends of the number of AG's and log buffers. JFS is by far the less cpu intensive, But on today's machines hard drive is the bottleneck. Plus, I rather by far real world test in the spirit of what I could myself do, than random benchmark. |
Maybe it's not *that* much, but it's quite a bit more than other filesystems . However, the benchmark might be taking that into account. I agree, a real world test would be much more important.
Yeah, hard drives are so much the bottleneck. But even processors are a bottleneck compared to the speeds that can be reached over networks (using Fiber, basically goes the speed of light). If hard drives have advanced as much as networks in the same time, there'd be hard drives that exist today which would have a petabyte of storage and transfer 500 GB per second. _________________ Carlton Stedman
Gentoo Metalheads on Last.fm: http://www.last.fm/group/Gentoo+Metalheads |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|