Date: Mon, 13 Sep 2004 11:42:11 +0200 From: =?ISO-8859-1?Q?S=F8ren_Schmidt?= <sos@DeepCore.dk> To: Frode Nordahl <frode@nordahl.net> Cc: current@freebsd.org Subject: Re: Promise PDC20267 ATA RAID, poor write performance Message-ID: <41456B73.9000701@DeepCore.dk> In-Reply-To: <B09A569C-0566-11D9-AA63-000A95A9A574@nordahl.net> References: <EC95929E-031F-11D9-AA63-000A95A9A574@nordahl.net> <4142D833.4060802@DeepCore.dk> <B09A569C-0566-11D9-AA63-000A95A9A574@nordahl.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Frode Nordahl wrote: > On Sep 11, 2004, at 12:49, S=F8ren Schmidt wrote: >> pizzabox# dd if=3D/dev/zero of=3Dfill bs=3D1m count=3D2000 >> 2000+0 records in >> 2000+0 records out >> 2097152000 bytes transferred in 122.908771 secs (17062672 bytes/sec) >=20 > I have done some more test now, reseated cables, disabling other IDE=20 > controllers etc. to no avail. I have tested both drives against the ICH= 2=20 > controller with very good results. >=20 > After some fooling around I found the following: > ar0: ATA RAID1 subdisks: ad4 ad6 status: READY > # dd if=3D/dev/zero of=3Dfill bs=3D1m count=3D100 > 100+0 records in > 100+0 records out > 104857600 bytes transferred in 29.298259 secs (3578970 bytes/sec) > # atacontrol detach ata2 > # dd if=3D/dev/zero of=3Dfill bs=3D1m count=3D100 > 100+0 records in > 100+0 records out > 104857600 bytes transferred in 3.080246 secs (34041957 bytes/sec) >=20 > (this also works when writing 2GB...) >=20 > That is, whenever I run with a single disk or a degraded RAID transfer = > speeds are ok. I have tried this both ways, so it is not one of the=20 > disks / cables that is broken. As soon as I rebuild the RAID again,=20 > write speeds drop to 3MB/s. >=20 > Any ideas as of how to debug this? Sounds strange indeed, there is some performance degradation to be=20 expected from RAID1 (you have to write to both disks) but nothing that=20 ruins performance this badly. I cannot reproduce it no matter what I=20 try, so I'm pretty sure the reason should be found outside ATA. It could be because the two channels are competing for the PCI bus and=20 the arbitration of that fails so you get alot of PCI retries, that is=20 outside ATA's reach to control.. -S=F8ren
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?41456B73.9000701>