> # tiobench.pl --numruns 5 --size 1024
> Size is MB, BlkSz is Bytes, Read and Write are MB/sec, Seeks are Seeks/sec
>
> Dir Size BlkSz Thr# Read (CPU%) Write (CPU%) Seeks (CPU%)
> ----- ------ ------- ---- ------------- -------------- --------------
> . 1024 4096 1 25.6001 12.0% 8.64345 7.08% 159.776 0.48%
> . 1024 4096 2 26.0927 12.3% 8.63389 7.11% 199.146 0.64%-W
> . 1024 4096 4 25.5174 12.0% 8.64470 7.19% 236.201 0.75%-W
> . 1024 4096 8 25.1011 12.1% 8.61819 7.25% 265.246 0.92%W
You know what? The most remarkable thing is how well Software-RAID holds
up to these numbers. My CPU load (on RAID1, granted, so no checksumming is
being performed) is not very bad in comparison (similar machine, only SW
RAID1 - check the benchmarks) - it is actually lower for reads, and only
about 1.5x for writes, considering I've got to write all data to both
disks - no write paralelism at all. Very interesting. I'd like to see how
it'd fare with better drives and perhaps two controllers.
Anyone have Soft-R5 numbers to compare to these?
[We've got hotswapping, boot/root-raid and quite proven stability - not
bad at all, considered the price tag :-) .]
Cheers,
--
_/\ Christian Reis is sometimes [EMAIL PROTECTED]
\/~ suicide architect | free software advocate | mountain biker
