[3871] in linux-scsi channel archive
Re: [BACK ON TOPIC] SCSI & IDE
daemon@ATHENA.MIT.EDU (Chris Wedgwood)
Fri May 8 05:16:14 1998
Date: Fri, 8 May 1998 21:13:21 +1200
From: Chris Wedgwood <chris@cybernet.co.nz>
To: Mr M S Aitchison <physmsa@cantua.canterbury.ac.nz>,
linux-smp@vger.rutgers.edu
Cc: linux-scsi@vger.rutgers.edu
In-Reply-To: <199805072251.KAA27297@cantua.canterbury.ac.nz>; from Mr M S Aitchison on Fri, May 08, 1998 at 10:51:20AM +1200
> Suppose you start with a single *IDE disk, fast and cheap, on an SMP
> Linux system you assume won't need much file access 'cause it is going
> to be number crunching all day. Then you need more disk
> capacity/performance, so you add SCSI. Assuming you use hdparm for
> best IDE performance will it still drag down the performance of the
> system?
I have multiple SCSI buses with several drives on them aswell as IDE.
Generally I find IDE is as fast as SCSI for all but really heavy loads,
although it eats way more CPU (DMA IDE).
I do however find seek times on SCSI the SCSI drives is much better (but
they cost a lot more), I also find under heavy load IDE gives me lots of
message like:
hda: irq timeout: status=0xd0 { Busy }
ide0: reset: success
Which really sux (IDE IO stops for about 2s).
> I suppose it depends on what the EIDE is used for. If it is just "system
> files" (/var, /usr, etc) or just swap, could it impair CPU performance
> (given the Linux 2.0 system call "bottleneck" especially)??
>
> Would you REMOVE the IDE?
No.
I'm also not sure I'd call the linux 2.0 syscall a bottle neck for lots of
disk IO. Generally I find disk seeks time to be the bottle neck.
-Chris
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
