|
About ::
TODO ::
Blog ::
RSS ::
Old blog ::
Projects ::
GIT ::
Gallery ::
Notes
Fri, 30 May 2008
Local filesystem randomg read/write performance. POHMELFS parallel testing.
scsi0 : Adaptec AIC79XX PCI-X SCSI HBA DRIVER, Rev 3.0
<Adaptec AIC7902 Ultra320 SCSI adapter>
aic7902: Ultra320 Wide Channel A, SCSI Id=7, PCI-X 67-100Mhz, 512 SCBs
scsi 0:0:2:0: Direct-Access SEAGATE ST3300007LC 0003 PQ: 0 ANSI: 3
target0:0:2: asynchronous
scsi0:A:2:0: Tagged Queuing enabled. Depth 32
target0:0:2: Beginning Domain Validation
target0:0:2: wide asynchronous
target0:0:2: FAST-160 WIDE SCSI 320.0 MB/s DT IU QAS RDSTRM RTI WRFLOW PCOMP (6.25 ns, offset 63)
target0:0:2: Ending Domain Validation
Kernel version is 2.6.25 (and 2.6.24 for the first ext3 test).I used two such machines as servers for iozone read/reread, write/rewrite and random read/write testing. File size is limited to 8Gb only, since it is the only interesting fair case, record size varies from 8Kb to 1Mb. Before I started 8Gb POHMELFS testing, I decided to check how local filesystem behave in such scenario. XFS was tuned this way: ( mkfs.xfs -d agcount=75 -l size=64m /dev/sdc1;
mount -o logbufs=8,nobarrier,noatime,nodiratime,osyncisdsync /dev/sdc1 /mnt/)Ext3 was created and mounted with default options on machine with only 4Gb of RAM though. So, testing. Here is a results table from iozone (before I interrupted it) with read/reread, write/rewrite and random read/write tests for XFS (either default, or tuned like on link above).
random random
KB reclen write rewrite read reread read write
8388608 8 73671 64052 77565 80107 35281 5085
8388608 16 74437 66095 77611 80065 66854 8808
8388608 32 74683 66780 77564 80202 121442 14576
8388608 64 74936 66908 77537 80372 215377 22583
8388608 128 74928 68598 77542 80247 339304 32280
8388608 256 73609 69615 77534 80143 365081 40571
8388608 512 73763 69830 77547 80317 420704 48501
8388608 1024 73940 69474 77602 80065 406266 47295
I.e. 5 MB/s random write speed for 8kb record!Do you really want to know ext3 speed? Pregnant kids and women should skip next paragraph. I interrupted test after almost 2 (!) hours or random writing of 8Gb file with 8Kb records on default ext3. Test was not completed and I do not really know its performance (note, that this machine has only 4Gb of ram, other hardware details were described above), but it will be less than 1 MB/s. Ext4 behaves much better in this aspect (ount options: rw,noatime,data=writeback,extents):
random random
KB reclen write rewrite read reread read write
8388608 8 69593 74200 77324 81340 35538 5088
8388608 16 66745 70038 73676 77271 65715 8704
8388608 32 68253 70320 73652 77258 121690 14469
8388608 64 68421 71291 73653 77042 209629 22005
8388608 128 68438 71340 73658 76988 332021 30381
8388608 256 68921 71254 73651 76912 435586 40683
8388608 512 69079 71728 73551 76815 549136 49298
8388608 1024 66611 71217 73683 76581 552459 49220
POHMELFS results are coming...
/devel/fs :: Link / Comments (0) Please solve this captcha to be allowed to post (need to reload in a minute): 67 - 96 Comments are closed for this story. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||