Repost: Bug in RAID subsys and/or cache handeling

From: Roy Sigurd Karlsbakk (roy@karlsbakk.net)
Date: Sun Dec 16 2001 - 12:30:25 EST


hi all

I'm sorry to repost this - I just need some sort of fix, or perhaps try
another OS, like a BSD or something

On my computer, I've set up a 2-disk stripe set (RAID-0). Each disk has
some 120GB disk space, and is pretty fast. Each disk can do some 40 megs
per second disk-to-memory.

However, when reading multiple, large files (3 gigs each) simultanously,
vmstat first shows quite high throughput (~ 40-50 megs per sec). But as I
reach the point where I've filled the cache, speed is reduced to 1-2 megs
per sec.

I'm running 2.4.16 with the latest tux patches plus Andrea's IDE patch.
I'm positive that this is not a tux problem. I've also tried 2.4.17-rc1,
but all I got out of that was some 20-30% slower I/O.

Hardware:

Athlon 1133MHz
1GB RAM (highmem disabled in kernel)
1x20 gig disk for root partition
2x120 gig disks in RAID-0 (chunk size = 4096) for data partition

Please help me out here...

Regards

roy

--
Roy Sigurd Karlsbakk, MCSE, MCNE, CLS, LCA

Computers are like air conditioners. They stop working when you open Windows.

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Sun Dec 23 2001 - 21:00:11 EST