Date: Thu, 28 Mar 2002 16:37:40 -0700 From: "Kenneth D. Merry" <ken@kdm.org> To: Lukas Ertl <l.ertl@univie.ac.at> Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: ciss driver and tagged queuing Message-ID: <20020328163740.A24032@panzer.kdm.org> In-Reply-To: <20020328232901.L214-100000@korben.in.tern>; from l.ertl@univie.ac.at on Thu, Mar 28, 2002 at 11:35:01PM %2B0100 References: <20020328152644.A23074@panzer.kdm.org> <20020328232901.L214-100000@korben.in.tern>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Mar 28, 2002 at 23:35:01 +0100, Lukas Ertl wrote: > On Thu, 28 Mar 2002, Kenneth D. Merry wrote: > > > What RAID level are you using? There are significant differences (on the > > smart array controller at least) between the block read and write results. > > With the SCSI->IDE controller, oddly, the write results are faster than the > > read results. > > RAID5. I don't have a lot of experience with RAID, so I don't know if the > number of disks have an influence on the result (12 80GB IDE disks in the > SCSI/IDE array, 8 72GB disks in the Compaq array). That explains why the block read performance is faster than the write performance on the Smart Array controller. With RAID-5, you have to do parity calculation on writes, so they'll be somewhat slower, more so if the board doesn't have a hardware parity engine. That doesn't explain why writes are faster than reads on the other array. > > What does dmesg say about the card? From looking at the driver, it looks > > like the "max outstanding commands" value that is printed out on boot will > > tell us how many outstanding transactions the card claims to support. > > Ok: > > ciss1: <Compaq Smart Array 5300> port 0x3000-0x30ff mem > 0xf7e00000-0xf7efffff,0x > f7fc0000-0xf7ffffff irq 3 at device 4.0 on pci7 > ciss1: using 256 of 1024 available commands > ciss1: 1 logical drive configured > ciss1: firmware 1.76 > ciss1: 2 SCSI channels > ciss1: signature 'CISS' > ciss1: valence 1 > ciss1: supported I/O methods 0xe<simple,performant,MEMQ> > ciss1: active I/O method 0x3<simple> > ciss1: 4G page base 0x00000000 > ciss1: interrupt coalesce delay 1000us > ciss1: interrupt coalesce count 0 > ciss1: max outstanding commands 1024 > ciss1: bus types 0x2<ultra3> > ciss1: server name '' > ciss1: heartbeat 0x20000022 > ciss1: 1 logical drive > ciss1: logical drive 0: RAID 5, 416768MB online Okay, it looks like you should be able to send at least 256 simultaneous commands. What does dmesg say about da1 (which I assume is the smart array device)? (might be faster to just send the full dmesg output) Ken -- Kenneth Merry ken@kdm.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020328163740.A24032>