raid10 far layout outperforms offset at writing? (was: Help with chunksize on raid10 -p o3 array)
I found this old message:
Quoted 25 lines
Peter Rabbitson
Mon, 19 Mar 2007 06:14:38 -0800Peter Rabbitson wrote:
I have been trying to figure out the best chunk size for raid10
before migrating my server to it (currently raid1). I am looking at 3 offset stripes, as I want to have two drive failure redundancy, and offset striping is said to have the best write performance, with read performance equal to far.Incorporating suggestions from previous posts (thank you everyone), I used this modified script at http://rabbit.us/pool/misc/raid_test2.txt
To negate effects of caching memory was jammed below 200mb free by using a full tmpfs mount with no swap. Here is what I got with far layout (-p f3): http://rabbit.us/pool/misc/raid_far.html The clear winner is 1M chunks, and is very consistent at any block size. I was surprised even more to see that my read speed was identical to that of a raid0 getting near the _maximum_ physical speed of 4 drives (roughly 55MB sustained across 1.2G). Unlike offset layout, far really shines at reading stuff back. The write speed did not suffer noticeably compared to offset striping. Here are the results (-p o3) for comparison:
http://rabbit.us/pool/misc/raid_offset.html, and they roughly seem to correlate with my earlier testing using dd.
Quoted 7 lines
So I guess the way to go for this system will be f3, although the md(4) says that offset layout should be more beneficial. Is there anything I missed while setting my o3 array, so that I got worse performance for both read and write compared to f3?
Once again thanks everyone for the help.
Peter
The links were not valid anymore. I wanted to see the results and possibly include the results in the performance wiki page I would appreciate some new links here.
Furthermore some comments to the post: My take on o3 vs f3 is that both in theory and practice f3 should be much faster for sequential reading, as the layout is equivalent to raid0. For random reading and sequential and random writing f3 and o3 (and the same goes for the more normal f2 vs o2) should be about the same, especially when a filesystem and its associated elevator algorithm is employed.
Best regards
keld
Quoted 9 lines · Keld Jørn Simonsen
I found this old message:
Quoted 3 lines
Peter Rabbitson
Mon, 19 Mar 2007 06:14:38 -0800The links were not valid anymore. I wanted to see the results and possibly include the results in the performance wiki page I would appreciate some new links here.
I apologize, I don't have the data available anymore.
Quoted 6 lines
Furthermore some comments to the post: My take on o3 vs f3 is that both in theory and practice f3 should be much faster for sequential reading, as the layout is equivalent to raid0. For random reading and sequential and random writing f3 and o3 (and the same goes for the more normal f2 vs o2) should be about the same, especially when a filesystem and its associated elevator algorithm is employed.
Yes, this is what I also concluded since I wrote this email. I am in the process of upgrading my raid setup, and while I am at it I am leaving 5GB blank partitions at the start of all my workstations spindles, so I can get some real testing at night. I will share my methodology with the list before I commence testing (which should take about 20 days the way I am planing it).
But first comes the vacation - happy holidays to you too guys. --
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 7 lines · Peter Rabbitson
Quoted 6 lines · Keld Jørn Simonsen
I found this old message:
Quoted 3 lines
Peter Rabbitson
Mon, 19 Mar 2007 06:14:38 -0800
Quoted 5 lines
Quoted 3 lines
The links were not valid anymore. I wanted to see the results and possibly include the results in the performance wiki page I would appreciate some new links here.
I apologize, I don't have the data available anymore.
'
OK.
Quoted 7 lines
Quoted 6 lines
Furthermore some comments to the post: My take on o3 vs f3 is that both in theory and practice f3 should be much faster for sequential reading, as the layout is equivalent to raid0. For random reading and sequential and random writing f3 and o3 (and the same goes for the more normal f2 vs o2) should be about the same, especially when a filesystem and its associated elevator algorithm is employed.
Quoted 6 lines
Yes, this is what I also concluded since I wrote this email. I am in the process of upgrading my raid setup, and while I am at it I am leaving 5GB blank partitions at the start of all my workstations spindles, so I can get some real testing at night. I will share my methodology with the list before I commence testing (which should take about 20 days the way I am planing it).
I have tried to persuade Neil to change the description for MD to reflect the above, but until now with no luck.
I look forward to see your new tests!
Quoted 1 line
But first comes the vacation - happy holidays to you too guys.
Yes, happy holidays to all!
Best regards
Keld
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html