On Tue Jul 20 23:48:17 2004, Dan Nelson wrote: > In the last episode (Jul 21), jesk said: > > i figured out that the performanceloss only really occur if the > > process is heavily writing on the filesystem. dd if=/dev/zero > > of=/dev/null bs=128k doesnt hurt much in responsetime of parallel > > processes, but when dd operates on the filesystem with of=foo every > > process will be affect in executiontime. a simple ps or ls meanwhile > > dding onto the disk will be hang for dozen of seconds. > > Ah. now that's a different story. You're out of the control of the > process scheduler and into the disk. I don't suppose you're using an > IDE/ATA disk with no tagged queueing? :) Run "dmesg | grep depth.queue" > to see how many requests can be queued up on your disk at once. > > That dd is stuffing lots of dirty data into the disk cache, and all the > other processes have to wait in line to get their I/Os done. You'll > see much better results from a SCSI disk (with usual queue depths > between 32 and 64), and even better results from a multi-disk hardware > RAID array (which will have a large write cache). Same here. I tried both with IDE and SCSI. With IDE it's terrible, ps for example takes seconds to execute. With an 2 cpu machine with scsi hdd's, it's better but there's some delay, too. I tested on a FreeBSD 4.9 machine with ide disks and there is no delay, it works perfect. That makes FreeBSD-5 really unusable for me. Best regards, Michael Riexinger systems engineer -- claranet gmbh internet service provider tel +49 (0) 69 - 40 80 18 - 300 email: michael.riexinger_at_de.clara.net http://www.claranet.de/Received on Wed Jul 21 2004 - 06:23:40 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:02 UTC