Subject: Re: 5000/240 and SCSI
To: Simon Burge <email@example.com>
From: Jonathan Stone <jonathan@DSG.Stanford.EDU>
Date: 03/17/1999 08:25:07
In message <199903171304.AAA28073@balrog.supp.cpr.itg.telecom.com.au>,
Simon Burge writes:
>Note that the 5000/2xx series machine have a 25MHz
>TURBOchannel, unlike the /xx and /1xx machines have.
>Some figures for "time dd if=/dev/rrzXc of=/dev/null bs=32k count=32k".
>All disks are RZ28Ms.
Eek. First time i tried thsi the cut-and-paste lost that last 'k'....
>5000/240 NetBSD 1.3.3 onboard SCSI 0.598u 31.750s 4:44.83 11.3% 3.60 MB/s
>5000/240 NetBSD 1.3.3 T/C SCSI 0.851u 30.421s 7:17.81 7.1% 2.34 MB/s
>5000/260 NetBSD 1.3K onboard SCSI 0.392u 13.416s 4:19.68 5.3% 3.94 MB/s
>5000/260 Ultrix 4.5 onboard SCSI 0.855u 10.307s 5:42.74 3.2% 2.99 MB/s
>5000/260 Ultrix 4.5 T/C SCSI 0.703u 11.878s 5:45.86 3.6% 2.96 MB/s
on a 5000/150 with an (oldish) IBM DCAS-34330 drive:
5000/150 NetBSD 1.3K ioasic SCSI 0.377u 19.547s 5:13.13 6.3% 3.43 MB/s
It looks like some kernel profiling is in order.
``Your mission, should you choose to accept it...''
And those system times on the /240 seem high compared to the 260.
Does that reflect real improvements since 1.3.3?
>Given that TURBOchannel is a 32bit bus, you'd think that a 25MHz bus
>with a peak bandwidth of 100Mbyte/s could sustain more than the 5MB/s we
>can get out of narrow 8bit SCSI. Ultrix's figures of almost identical
>figures for onboard and T/C give some hope of getting better figures for
>NetBSD's T/C SCSI performance.
Well, around 5 MB/sec TCP throughput is all I could squeeze out of a
couple of different high-speed NICs, due to CPU bandwidth limits. In
fact, a 5000/200 got slightly higher bandwidth than a /240. Which
turned out to be due to the memory stalls from synching a 40MHz CPU to
a memory/TC subsystem running at 25MHz.
>As far as part numbers, Ultrix's tc_option_data.c lists a PMAZ-AA as an
>"asc" board and "PMAZB-AA" and "PMAZC-AA" as using the "tcds" driver.
>The on-board controller probes as a "PMAZ-BA" under Ultrix.
They're making it up as they go along, at least vis-a-vis the official
TC specs. There's no acutal "PMAZ-BA " string in the PROM. But it's
hard to argue with DEC about what DEC devices are called :).