20 Mar
2002
20 Mar
'02
12:41 p.m.
On Wed, 20 Mar 2002, Gunther Schadow wrote:
So I looked at statistics. Nothing paged out (0.5 GB
of RAM),
but looks like the system is heavily I/O bound and I/O
is quite slow. I even distributed disk load over 3 RA90, though all on
the same controller, but at least that's a KDM70, directly on the XMI
bus. Still, disk access seems just a bit on the slow side. Why?
The RA90 is not a speed demon at 3600 rpm and SDI seems to max out under
3MB/s.
RA90, fixed-media 9" DSA disk drive
Capacity ... 1.216 GB, 2376153 total blocks
2,649 cylinders, 13 tracks/cylinder, 69 sectors
peak xfer rate ........... 2.77 MB/s
spiral xfer rate ......... 1.72 MB/s
average access time ..... 26.8 msec
average seek time ....... 18.5 msec
track-track seek time .... 5.5 msec
head switch time ......... 3.0 msec
average latency .......... 8.33 msec
rotational speed ...... 3600 RPM
request rate ............ 39 I/Os per second at 50 msec response time
46 I/Os per second at 100 msec response time
It's interesting. At some time soon, may be early
this summer,
I'll give a VAX party where one of the highlights will be a
race between my i486/33 and the 6460 in compiling something,
there we can see if it's really just my perception that the
6460 is kind of slow for the price, even measured by past
standards.
Wouldn't the 65xx or 66xx be more around the time of a 486? I don't know
when the 64xx came out but I thought it was earlier. Are you planning to
run a *bsd on the 486 from the same timeframe as Ultrix 4.5 or one with a
scheduler and interrupt handler potentially more recently optimized?
Does your i486 feature long ribbon cable to dual ported IDE drives to
support its clustering? If not, perhaps you were paying for features
other than speed, ones that Ultrix took less advantage of than VMS.