Solved my issue with LSI 8480E without BBU. With "write: BadBBU", "cache: enabled", and "io: cached", performance rose to around 200MB/s from 20MB/s.
With BadBBU/enabled cache/io cached: [root@ ~/bench]# ./bench /dev/mfid0p1 65536 file /dev/mfid0p1, bs 65536 rate: 228130816 Bps, write: 285 us rate: 206635008 Bps, write: 299 us rate: 206372864 Bps, write: 305 us rate: 206438400 Bps, write: 307 us ... With Write through: [root@ ~/bench]# ./bench /dev/mfid0p1 65536 file /dev/mfid0p1, bs 65536 rate: 40763392 Bps, write: 1609 us rate: 19202048 Bps, write: 2260 us rate: 25231360 Bps, write: 2301 us rate: 19660800 Bps, write: 2418 us ... 10x performance win, so this is one thing off my back. Kind regards, Fredrik Widlund Fredrik Widlund wrote: > 256MB cache, no BBU. Tried a lot of different combinations of settings. > > fbsd 6.2pre writes 220MB/s with raid-0 > > If I boot windows server 2003 instead, it writes at around 180MB/s with > raid-5 (same configs). > > With fbsd6.2pre, I get the "best" performance with BadBBU/direct > io/disabled cache and 8 drives, maybe around 50 MB/s, however I get a > zig-zag performance pattern, with the adapter running at 500% (gstat) > utilization one second, then down to 0% for maybe 3 seconds, back, etc. > > Fredrik > > Ivan Voras wrote: > >> Fredrik Widlund wrote: >> >>> Ivan Voras wrote: >>> >>>> Several: >>>> >>>> - are there cache differences between the controllers (amount of >>>> memory, cache policy)? >>>> >>> Default settings on both. >>> >> Maybe you should check what the defaults are :) Especially the amount >> of memory and is there a battery to back the cache. >> >> >>>> - how does writing directly to the device (bypassing file system) >>>> compare? >>>> >>> Drives are four seagate 7200.10 400GB in a Raid-5 configuration. >>> >>> [/mnt/test (/dev/mfid0p1 mounted)] >>> read: 200MB/s >>> write: 15MB/s >>> >>> [/dev/mfid0p1] >>> read: 200MB/s >>> write: 8MB/s >>> >>> [/dev/mfid0] >>> read: 200MB/s >>> write: 10MB/s >>> >> This is bad :( The difference between p0 and raw device might indicate >> a stripe size misalignment, but the values are too low in any case. >> >> >> _______________________________________________ >> [email protected] mailing list >> http://lists.freebsd.org/mailman/listinfo/freebsd-stable >> To unsubscribe, send any mail to "[EMAIL PROTECTED]" >> > > _______________________________________________ > [email protected] mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "[EMAIL PROTECTED]" > _______________________________________________ [email protected] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
