Re: ciss(4): speed degradation for Compaq Smart Array [edited]

From: Martin Nilsson <martin_at_gneto.com>
Date: Wed, 30 Mar 2005 22:03:32 +0200
Doug White wrote / skrev:
> 
> I'd also suggest using a tool like iozone to run your tests instead of dd.
> Unless your workload consists of entirely sequential writes this perf test
> is worthless.

What is wrong with using dd to measure max read/write bandwith of the 
controller-disk system?

I use the following to determine if there are any obvious problems with 
new hardware: (ad1 should not have any useful data).
At the same time I run "systat -v 1" on another console.

dd if=/dev/zero of=/dev/ad1 bs=128k
dd if=/dev/ad1 of=/dev/null bs=128k

On a modern disk this should give numbers in the vicinity of 50MB/s at 
the beginning of the disk. This tells me in seconds if there is 
something wrong with the controller, cable, disk (or FreeBSD driver).

If I can't get decent bandwith with this simple test, how could I get 
good performance with any other test program?


Regards,
Martin
Received on Wed Mar 30 2005 - 18:03:36 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:30 UTC