In article <[EMAIL PROTECTED]> you write:
>
> -------Sequential Output-------- ---Sequential Input-- --Random--
> -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
>Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
> 1024 4799 23.8 5053 4.9 3611 9.6 12344 50.5 14878 22.8 186.6 2.0
> 1900 5194 25.8 5048 4.9 3506 9.4 12595 51.7 14682 22.7 134.5 1.5
>
>This is in a RAID-5 config with 5 9gb UW drives in a dual PII-450 with
>512mb RAM running kernel 2.2.2-smp.
Here are a few results using the latest software RAID patches on
2.0.36. I'm using an Intel N440BX motherboard with one P2/350, 256 MB
RAM, Symbios 875/876 UW SCSI controllers, and 5 Seagate ST39173W
drives. All tests use a 4k e2fs blocksize and a 32k RAID chunk size.
This is SW RAID5, with all drives on a single UW SCSI channel:
-------Sequential Output-------- ---Sequential Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
512 12309 78.7 26745 29.1 7949 21.1 14941 76.4 24520 24.0 417.7 5.0
512 12449 79.6 26387 28.2 8068 21.0 15466 79.0 24476 24.0 467.9 6.6
512 11316 72.1 26231 27.3 7976 21.2 15153 77.7 24501 23.2 438.4 6.5
This is SW RAID5, with 2 drives on one channel and 3 on another.
Notice that block I/O is up ~5MB/sec, and rewrite is up 1.5 MB/sec.
-------Sequential Output-------- ---Sequential Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
512 11946 79.4 31607 37.2 10025 26.5 15105 77.8 28394 26.5 425.1 7.1
512 12502 80.8 31837 34.8 9641 25.6 15970 81.4 28779 27.5 472.6 5.3
512 12555 79.9 31952 35.5 9556 26.0 16128 82.8 28801 28.3 495.8 6.9
This is SW RAID0, in the same configuration as the last test, just for
comparison. Notice the drop in random seeks on the last test.
-------Sequential Output-------- ---Sequential Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
512 15077 96.9 48882 62.5 14713 40.5 18463 94.1 42459 49.6 486.9 8.2
512 15119 96.9 47023 58.5 14563 40.8 17382 88.7 42721 49.1 472.2 8.0
1536 15047 97.0 45330 59.5 15316 42.1 18905 96.6 40239 47.7 323.2 7.8
Finally, just for fun, here's a 4-drive RAID0 array on a single UW
controller. Notice the block write speed -- this is almost complete
utilization of a UW SCSI bus. The NCR53c8xx driver looks pretty good
:-).
-------Sequential Output-------- ---Sequential Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
512 2632 98.6 37529 45.6 12186 39.4 17222 88.2 24152 46.5 458.2 7.8
I'm planning on adding 6 more drives (and two more controllers) into
the array this afternoon, just to see how fast this hardware can
actually move data. I'm rather impressed so far.
Scott
-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/mentre/smp-faq/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]
