|
XFS drops create/delete files to 6.6% of EXT3 (sw raid) and to 0.6% (hw raid)
I am preparing a new server, and benchmarking EXT3 against XFS, both using software RAID and hardware RAID using a 3ware 9650SE-4LPML.
Using bonnie++ as a benchmark, I am seeing significant performance boosts in my block sequential reads and writes moving from EXT3 to XFS. I am aware that XFS won't create and delete files as quickly as EXT3, however I am seeing drops from 29455/second to 1957/second using software RAID, and from 32524/second to 189/second using hardware RAID. I'm not sure if when using software raid, if creating and deleting files should drop to 6.6% of EXT3. But, what I'm pretty sure of, is when using hardware raid, that creating and deleting files shouldn't drop to 0.6% of EXT3.
When using the 3ware card, mkfs.xfs defaulted to "data sunit=0 swidth=0 blks", which made me think that might be the problem, so I tried "-d sunit=128,swidth=384" with no effect. My 3ware card is using 64k stripes, so I calculated 64k stripes / 512 bytes = sunit 128, and multiplied by 3 drives (3 usable in a 4 drive raid 5) to get swidth 384. This is my first time calculating these values, so I'm not sure if I did this right. Unfortunately, these settings had no performance change.
If anyone could give me some pointers, I would much appreciate it!
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
SW RAID 5 EXT3___8G 54852 96 116904 37 52572 15 63851 96 189073 25 352.8 1
SW RAID 5 XFS____8G 61281 97 146610 32 48509 17 57977 89 180168 26 479.4 0
HW RAID 5 EXT3___8G 53483 93 108314 31 57674 12 62134 91 227582 19 329.9 0
HW RAID 5 XFS____8G 62518 99 211622 33 78473 12 68182 99 214218 15 484.6 0
------Sequential Create------ --------Random Create--------
-Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
SW RAID 5 EXT3___16 29455 89 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
SW RAID 5 XFS____16 _1597 11 +++++ +++ _1348 __7 _1398 _10 +++++ +++ __954 __5
HW RAID 5 EXT3___16 32524 95 +++++ +++ +++++ +++ +++++ +++ +++++ +++ +++++ +++
HW RAID 5 XFS____16 __189 _1 +++++ +++ __172 __0 __186 __1 +++++ +++ __122 __0
To create the software RAID 5, I am running "/sbin/mdadm --create /dev/md0 --verbose --level=raid5 --raid-devices=4 /dev/sd{b,c,d,e}".
To create the hardware RAID 5, I am using a 3ware 9650SE-4LMPL. I am using default values (64k stripe), except turning write cache on and setting StorSave to balance.
To create an ext3 filesystem, I am running "/sbin/mke2fs -j /dev/[md0/sdb1]".
To create a xfs filesystem, I am running "/sbin/mkfs.xfs -i size=512 /dev/[md0/sdb1]". Here's the output I get:
meta-data=/dev/sdb1______________isize=512____agcount=32, agsize=11443802 blks
_________=_______________________sectsz=512___attr=0
data_____=_______________________bsize=4096___blocks=366201664, imaxpct=25
_________=_______________________sunit=0______swidth=0 blks, unwritten=1
naming___=version 2______________bsize=4096
log______=internal log___________bsize=4096___blocks=32768, version=1
_________=_______________________sectsz=512___sunit=0 blks, lazy-count=0
realtime_=none___________________extsz=4096___blocks=0, rtextents=0
I also tried "/sbin/mkfs.xfs -i size=512 -d sunit=128,swidth=384 /dev/sdb1" and got the same results.
I also tried "/sbin/mkfs.xfs -d sunit=128,swidth=384 -l sunit=128 /dev/sdb1" and got the same results.
I attempted creating a larger logfile using "/sbin/mkfs.xfs -l size=256m /dev/sdb1", however I was told the maximum size was 128MB.
Last edited by darlingm; 04-23-2008 at 09:12 PM.
|
