gauging interest in VAX 6000-530

From: Allison J Parent <allisonp_at_world.std.com>
Date: Sun Oct 24 18:09:33 1999

<My main belief is that nobody is going to keep a VAX anything running with
<dozens of simultaneous users. So, if a VAX is to be something close

Very errorloaded belief. It's done, being done. In 1993 the style of
computing is actually quite different than current. large companies used
VAX or as400 as central computers and maybe also servers to those PC
users that wanted application autonomy.

<Now, perhaps if we were to port Apache to the VAX, and used that I/O bandwi
<on multiple DS3s, well, that's great.

Exists, I know of a company using a bunch of 6000s to service a user base
of some 1000 systems never mind running all the batch jobs like payroll,
job costing, database updates and other mundane stuff.

<Hey, I'm not saying the original IBM PC was going to outperform the VAX 650
<but a modern PC will crush any VAX in any application, IMHO, with equivalen
<h/w attached.

NO, simple reason, PCs don't have equivelent hardware.

<1) The names of these busses?

Some are FDDI, Massbus, CI, HSI, SCSI(I/II/IIW...), and a few properitory
busses.

<2) Their uses?

Two general catagories, memory interconnect and Storage interconnect.
There are several IO interconnects as well.

<3) Their peak and average throughputs?

I think FDDI is still in the 100++m/bytes/Sec region. Though there were a
few parallel busses that were 50-100mbytes/sec rate.

I will not argue this as PC are finally going fibre and Gigabit eithernet
but thats 10 years after the fact.

<I certainly know for a fact that UNIBUS performed very poorly. I don't hav
<data at my fingertips, but it seems to me it was around 10 Mb/s (that
<megabits/sec) peak throughput. [I prefer measuring throughputs in bits/se
<since that normalizes across different bus widths.]

Also loaded with errors. Systems process things in chunks. FYI the
pdp-11/70 was a 1mips (I think it was also 1mfps) machine despite that
slow unibus.

<Fast dual-port SRAM solves the problem, but commodity PCs aren't designed
<that way. Also, the AGP bus uses mega-RAM to speed up PC graphics, for exam

Exactly they arent designed that way. The cpus might do better in a real
system but even then there are throughput issues. Your original arguement
WAS a 486dx would kill the VAX 6000-530. the answer is still not ever!

In the end it still takes 10k$ to do what 10k$ of vax did then.

The rather specious arguement of my daddy can beat your son is is pointless.
I can easily beat a PDP-8 with a slow PDP-11, what have I proved other than
the next generation machine is better (wasnt that the point of its design?).
In 1993, PCs were plentyful but they were not the applications killers.
What they were then was the cheap workstations with 640x480x8 or maybe
800x600x8 graphics and often a lot less. Could a 486dx be made such that
it would be a crusher box... yes but, there would have been no software for
it as PC compatability was a must have back then other wise it was just
another soso cpu.

Allison
Received on Sun Oct 24 1999 - 18:09:33 BST