gauging interest in VAX 6000-530

From: Mike Cheponis <mac_at_Wireless.Com>
Date: Wed Oct 27 02:27:55 1999

> > real-life situations, one faster bus is BETTER than two separate busses.

> In the real world there are too many conditions on things to make general
> assumtions that always work.

Including this one, no doubt? ;-)


> For simplicity and cost as a generalization
> one fast bus will meet those needs and give good performance.

That statement is supported by fact!

> Now a counter to that. For error/fault tolerence, and if the fastest bus
> is not fast enough then two (or more) will always win but will not be
> cheaper.

I'm not sure I understand this: the fastest bus is not fast enough?

> I don't need numbers to prove that, the logic is deeper.

? Deep sub-micron or something?


> To tie that to
> the "old" machines they ran their generations PCI and then some. I'll bet
> in a year or two you make the exact same arguement about the obsolete PCI
> bus.

Gee, as much as folks like to pidgen-hole me, I'm really only trying to
understand the performance and lure of the old iron, like vaxes, in
relation to newer iron that I have or can use, and also understand better.

Call it "Comparative Computer Anatomy" if you will...

Perhaps people do not like my approach, but my aim is true.

>To make a machine I'd consider to be in the same class as the 6000
> was when new right now I'd expect to see all of the fastest busses in use
> and multiples of them for parallelism.

This is, unfortunatly, gobbdleygook to me. "multiples of them for parallelism"?
Why would this be necessary? Why can't one fast one be used?


> Current generation PCs most of us have access to are not this case and do
> perform well below the processor chips capabilities

I don't understand this, either. Are you saying that if there were
memory-speed peripherals, I/O could flow through the processor faster than
existing PC busses? If so, then I definitely agree with you. However,
then bus design gets complicated with cache-coherency etc. I think this
is why you want at least two busses on a machine: memory and I/O. (As
network speeds increase beyond 1 Gb/s, then perhaps a special Network Bus
will be required, too.)

> despite AGI, PCI,
> SCSI++ (LVUW), copper giga and fibre. Because even the simplest process
> like routing a packet requires the cpu to look at the address and see if
> it belongs to the port A or port B list and all the DMA in the world
> doenst get around that.

right! That's what I was arging before concerning "processing".


> A good example is the DEC ugly(by some) RQDX MSCP disk controller.
> While Qbus is maxed out around 4m words a second this controller can do
> one thing to keep the lowly vax off it. It can DO DMA from LINKED lists
> so that the controller is doing queued IOs. Now thats low end 1987
> MicrovaxII technology but if we scaled that and put it on the PC it could
> be keeping the cache filled and other tasks while all the cpu has to do is
> set up task lists for it.

Sounds like what we now call "scatter/gather DMA" - and is a great idea
for networking stacks.


> Another is this RAID thing, a good idea if you have SCSI. The VAX
> was doing it over 14 years ago as part of the OS (disk shadowing).

This is, as I understand it, not -quite- true. Disk Shadowing (what is
that, RAID-1 ?) is a simple technique, not as sophisticated as RAID-5,
for example.

But I remain unclear how this helps us understand busses?


> OK so your 486 can do interger math faster, few said it was false.

Right, and those that few that disagreed were wrong. ;-)

> It's
> just not enough of a measure. Less so to those guy where processing is
> putting the right data (or part) in the correct bin and charging off
> 10,223.245 (yes three digits!) each for them. It may be meaningless
> to the weather service that runs models that have data matricies that can
> fill gigabytes of ram (which has to be filled first!) from their terabyte
> disk farms.

Right. But, for my goal of attempting to put these old, "obsolete", err, I
mean "ancient", err, I guess it's PC (Politically Correct!) to say "Classic"
machines in perspective.


> All those MIPS, MFLOP and Dhrystones are wasted if the system structure
> is hung up waiting for the floppy.

You mean if the system is waiting for some I/O device? Sure!


> If your are preaching newer is faster, the choir is over there, they know
> the words.

I'm not. (But it -is- a fact, though. "Obsolete Ready" as they say...)

> If your saying PCs are better, well, put some bounds around
> that as I really feel it's one of those ALWAYS/NEVER generalizations
> that often bite.

For the nth time: a dx2/66 is 2 to 3x as fast as a uniprocessor 6500 on
Dhrystone 2.1; this is what I've maintained, and it helps put at least
part of the history of vaxes in perspective for me.

I think we are all fully aware that there are other aspects that need to
be considered to put the vax history in total perspective.

-Mike C

p.s. Allison, just curious, are you a recovering DECaholic? ;-) Your ferver
for the old DEC stuff is amazing to me! Religion? It's fun to watch!
Received on Wed Oct 27 1999 - 02:27:55 BST