On 2009-11-05 18:03, Jefferson Ogata wrote:
> On 2009-11-05 16:39, Brian A. Seklecki wrote:
>> Do tell. Post your results bonnie++ results somewhere? bon_csv2txt(1) or
>> html
Config as noted in my previous message, running bonnie++ -f against an
ext3 filesystem:
Using a linear LV on a single 7-disk RAID5:
Version 1.03e ------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
foo 32088M 139064 28 102222 16 337344 23
627.6 0
------Sequential Create------ --------Random
Create--------
-Create-- --Read--- -Delete-- -Create-- --Read---
-Delete--
files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
/sec %CP
16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
+++++ +++
foo,32088M,,,139064,28,102222,16,,,337344,23,627.6,0,16,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++
That's 139MB/s block write, 337MB/s block read.
Again, using a striped LV over 4 7-disk RAID5s:
Version 1.03e ------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
foo 32088M 314085 67 234891 39 583696 34
745.1 1
------Sequential Create------ --------Random
Create--------
-Create-- --Read--- -Delete-- -Create-- --Read---
-Delete--
files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
/sec %CP
16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
+++++ +++
foo,32088M,,,314085,67,234891,39,,,583696,34,745.1,1,16,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++
That's 314MB/s block write, 583MB/s block read.
Just for grins, to see what the box can do, once again, this time
striping an LV over 8 7-disk RAID5s using a second PERC6 (since the box
actually has another PERC6 with two more MD1000s attached to that):
Version 1.03e ------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
foo 32088M 385326 82 335313 58 971629 67
830.6 1
------Sequential Create------ --------Random
Create--------
-Create-- --Read--- -Delete-- -Create-- --Read---
-Delete--
files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
/sec %CP
16 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
+++++ +++
foo,32088M,,,385326,82,335313,58,,,971629,67,830.6,1,16,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++,+++++,+++
That's 385MB/s block write, 971MB/s block read--nearly 1GB/s--using 56
spindles in 4 enclosures, and two PERCs. This is using defaults across
the board and all stock software and firmware. And the box is doing
other stuff at the same time. Personally, I think that's not too shabby;
certainly it's possible the Areca cards are faster, but I really see
nothing to complain about here.
So, I'm really curious, under what configuration were you able to limit
a PERC6 to 30-50MB/s?
_______________________________________________
Linux-PowerEdge mailing list
[email protected]
https://lists.us.dell.com/mailman/listinfo/linux-poweredge
Please read the FAQ at http://lists.us.dell.com/faq
