Re: Soft raid 10

Well in case anyone is interested here is some basic bonnie output. I
ran it three times each, once on a single disk, once on raid 10, the
final time with the raid10 driver. They're all using XFS.

Version 1.03 ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
steven-desktop 2G 48843 97 78668 17 32771 6 43574 71 80993 6 255.6 0
steven-desktop 2G 47370 98 79139 17 33256 6 43666 70 81064 6 240.7 0
steven-desktop 2G 48741 97 78935 17 32162 6 43376 70 81060 6 255.9 0
steven-desktop 2G 46608 98 147768 28 35239 7 46790 84 156976
12 588.6 0
steven-desktop 2G 48686 98 144322 27 34038 7 46105 81 153161
11 613.3 1
steven-desktop 2G 49377 99 151909 29 32837 6 45848 81 154031
11 670.0 1
steven-desktop 2G 49436 99 143057 27 34825 6 49452 90 124514
9 648.6 1
steven-desktop 2G 48945 98 147303 30 33403 7 50411 92 133488
11 653.2 1
steven-desktop 2G 47680 98 148318 27 33958 7 49832 91 128711
10 613.9 1
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
files:max:min /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
steven-desktop 16 382 1 +++++ +++ 290 1 356 2 +++++ +++ 248 1
steven-desktop 16 372 1 +++++ +++ 295 1 354 2 +++++ +++ 264 1
steven-desktop 16 382 1 +++++ +++ 293 1 366 2 +++++ +++ 248 1
steven-desktop 16 4147 9 +++++ +++ 3173 5 3925 9 +++++ +++ 3055 6
steven-desktop 16 3977 9 +++++ +++ 3272 6 4225 10 +++++ +++ 3079 5
steven-desktop 16 4065 11 +++++ +++ 2951 4 4200 10 +++++ +++ 2716 5
steven-desktop 16 2201 0 +++++ +++ 1732 0 2380 0 +++++ +++ 1515 0
steven-desktop 16 2076 0 +++++ +++ 1795 0 1753 0 +++++ +++ 1750 0
steven-desktop 16 2126 0 +++++ +++ 1480 0 1350 0 +++++ +++ 1524 0

On 1/26/07, unit3 <unit3@demoni.ca> wrote:
> Steven Kurylo wrote:
> > We'll I'm using 4 external sata drives with an LSI SAS3442X.
> > Eventually I'm going to have to decide if we should get hardware raid
> > instead. Here is to hoping this will be good enough!
> I would suspect that if your machine is decently fast that the raid10 md
> driver should be faster than most single card hardware implementations,
> but I'll await your benchmarking. :)
>
> G
>
> --
> "Those who do not understand Unix are condemned to reinvent it, poorly."
> (Henry Spencer, 1987)
>
> To unsubscribe, send a message with the word "unsubscribe" (without the
> quotes) in the body to linux-request@slg.org
> Archives are at http://list.slg.org/
>
Received on Fri Jan 26 15:33:53 2007