Date: Tue, 31 Oct 2000 20:43:27 -0400 (AST) From: The Hermit Hacker <scrappy@hub.org> To: Tom Samplonius <tom@sdf.com> Cc: freebsd-scsi@freebsd.org Subject: Re: iostat: tps for SCSI drives ... Message-ID: <Pine.BSF.4.21.0010312020040.494-100000@thelab.hub.org> In-Reply-To: <Pine.BSF.4.05.10010310908560.24950-100000@misery.sdf.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 31 Oct 2000, Tom Samplonius wrote:
>
> On Tue, 31 Oct 2000, Marc G. Fournier wrote:
>
> > morning all ...
> >
> > what is considered to be a 'saturated drive', as far as tps is
> > concerned? I have a database server that I found one drive to be overused
> > with another not used at all, so I started moving around databases to
> > balance off the load a bit ... which helped alot. But am wondering if
> > there is an "acceptable tps level" for a drive before you should look at
> > moving things around ?
>
> Well, if you assume that each transfer requires a seek, and you know the
> average seek time of your drive, you'll be about to figure out the number
> of transfers per second your drive can do (assuming each transfer requires
> a seek). If you are using SCSI drives with tagged command queues, you can
> use camcontrol to see how many pending commands there are.
Okay, how do I read this:
pgsql# camcontrol tags da0 -v
(pass0:ahc0:0:0:0): dev_openings 41
(pass0:ahc0:0:0:0): dev_active 0
(pass0:ahc0:0:0:0): devq_openings 41
(pass0:ahc0:0:0:0): devq_queued 0
(pass0:ahc0:0:0:0): held 0
(pass0:ahc0:0:0:0): mintags 2
(pass0:ahc0:0:0:0): maxtags 255
pgsql# camcontrol tags da1 -v
(pass1:ahc0:0:1:0): dev_openings 1
(pass1:ahc0:0:1:0): dev_active 0
(pass1:ahc0:0:1:0): devq_openings 1
(pass1:ahc0:0:1:0): devq_queued 0
(pass1:ahc0:0:1:0): held 0
(pass1:ahc0:0:1:0): mintags 2
(pass1:ahc0:0:1:0): maxtags 255
pgsql# iostat -n5 -t da 5
tty da0 da1 da2 cpu
tin tout KB/t tps MB/s KB/t tps MB/s KB/t tps MB/s us ni sy in id
77083887 48.00 36 1.67 0.00 0 0.00 0.00 0 0.00 23 0 9 0 68
0 9856 42.97 34 1.45 0.00 0 0.00 8.07 23 0.18 0 0 0 0 0
0 9856 39.77 41 1.57 0.00 0 0.00 8.00 23 0.18 0 0 0 0 0
max/mintags aren't a problem ... I take it, from what I'm reading in the
man page, the 'pending' is the dev_active field? So, there is nothing
pending?
The two drives in use above are:
da0 at ahc0 bus 0 target 0 lun 0
da0: <SEAGATE ST39103LW 0002> Fixed Direct Access SCSI-2 device
da0: 80.000MB/s transfers (40.000MHz, offset 15, 16bit), Tagged Queueing Enabled
da0: 8683MB (17783240 512 byte sectors: 255H 63S/T 1106C)
da2 at ahc0 bus 0 target 2 lun 0
da2: <FUJITSU MAE3182LP 0112> Fixed Direct Access SCSI-2 device
da2: 80.000MB/s transfers (40.000MHz, offset 63, 16bit), Tagged Queueing Enabled
da2: 17431MB (35700480 512 byte sectors: 255H 63S/T 2222C)
So, for the Seagate, according to Seagate, the avg seek time is 5.2ms ...
so I should max out around 192.3 tps on that drive? I know, not set in
stone, but a reasonable number to work with, right? And, from what I can
tell about the Fujitso, they are running around 7ms, so ~142tps?
Does this make sense?
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?Pine.BSF.4.21.0010312020040.494-100000>