The Perfect Computer - 36 bits?
|
From: jmfbahciv on 29 Mar 2007 09:20 In article <571pauF2a7qasU1(a)mid.individual.net>, =?ISO-8859-1?Q?Jan_Vorbr=FCggen?= <jvorbrueggen(a)not-mediasec.de> wrote: >> It may not have helped you, but there have been a lot of occasions in >> which I have needed to know where something might have been updated >> NOT in the main modules. And there were a zillion reads and loads of >> the address :-( > >VMS has a WATCH facility which lets you monitor reads or writes of given >system addresses, and will report the places from where they are happening, in >symbolic form. Yes, that's really a nice thing, sometimes it even helps you to >diagnose those ugly race conditions. The PDP-10s provided address break. In the SMP Unix project that JMF worked on, this was done with software since the hardware wouldn't help. /BAH
From: jmfbahciv on 29 Mar 2007 09:21 In article <eugb7d$ric$1(a)gemini.csx.cam.ac.uk>, nmm1(a)cus.cam.ac.uk (Nick Maclaren) wrote: > >In article <571pauF2a7qasU1(a)mid.individual.net>, >=?ISO-8859-1?Q?Jan_Vorbr=FCggen?= <jvorbrueggen(a)not-mediasec.de> writes: >|> >|> > It may not have helped you, but there have been a lot of occasions in >|> > which I have needed to know where something might have been updated >|> > NOT in the main modules. And there were a zillion reads and loads of >|> > the address :-( >|> >|> VMS has a WATCH facility which lets you monitor reads or writes of given >|> system addresses, and will report the places from where they are happening, in >|> symbolic form. Yes, that's really a nice thing, sometimes it even helps you to >|> diagnose those ugly race conditions. > >As do many other systems. That can help, but didn't/doesn't help in the >cases I am thinking of. Would address break have helped in those cases? Assume you could define a mask and break on write. /BAH
From: jmfbahciv on 29 Mar 2007 09:22 In article <571ps8F2atkpuU2(a)mid.individual.net>, =?ISO-8859-1?Q?Jan_Vorbr=FCggen?= <jvorbrueggen(a)not-mediasec.de> wrote: > > Not having source *or* fiche made me very nervous. > >Ah yes, and now think about all those business- and safety-critical customers >of the Microsoft operating systems... > > Jan > >PS: For those OSes, quite likely even having the source wouldn't help you much... > But having sources would give a snowball in hell a chance to keep his water together. /BAH
From: Jan Vorbrüggen on 29 Mar 2007 09:34 > Neither would matter. Look if you increase your "CPU speed" by twice, > your system will then be constantly waiting on I/O becaues the > CPU got its job done faster. YOur system software and usage had > been tweaked over the years to accomodate the behaviour of a VAX > with its peripherals (this includes memory). Now you replace > the CENTRAL processing unit with something that goes twice as fast. If VAX processor speed had increased in parallel with the rest of the industry, that would be a valid comment that applied almost all around, but it did not. And I and many people I know are in the HPC department with regard to processor speed - heck, for a lot of codes I used we got practically linear speedup with processor MHz, main memory speed be damned. I/O speed - if not paging or swapping - was simply irrelevant. Even for Oracle, I/O isn't necessariyl the limiting factor. Jan
From: Jan Vorbrüggen on 29 Mar 2007 09:37
> And some of that work was done by JMF, the other half of my > username. It took those with TOPS-10 experience to cause VMS > to evolve to be an OS that was useful. (The work wasn't jsut > JMF's but the whole contingent of PDP-10 developers who moved > to the 32-bit world after the 36-bits were cancelled.) The distinguishing feature of VMS, IMNSHO, is the VMScluster, with the connection manager and the lock manager as the main supporting pieces of software of that capability. TOPS-10 or -20 never had anything like it. Where was Leslie Lamport before this happened? (A lot of the VMScluster stuff is based on his ideas.) Enterprise-level batch and print management, for instance, I can well believe was derived from, in particular, TOPS-20 - TOPS-10 was better than early versions of VMS in this respect, but VMS caught up after some time. Jan |