Re: Why was VAX abandonned ?
From: Paul Sture (nospam_at_sture.homeip.net)
Date: 02/10/04
- Next message: John Santos: "Re: Why was VAX abandonned ?"
- Previous message: JF Mezei: "Re: HP/Intel Itanium is dead, long life to HP/AMD Athlon"
- In reply to: Daryl Jones: "Re: Why was VAX abandonned ?"
- Next in thread: JF Mezei: "Re: Why was VAX abandonned ?"
- Reply: JF Mezei: "Re: Why was VAX abandonned ?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Tue, 10 Feb 2004 08:12:00 +0100
Daryl Jones wrote:
>
> glen herrmannsfeldt <gah@ugcs.caltech.edu> wrote in message news:<chFVb.208883$nt4.986971@attbi_s51>...
> > Daryl Jones wrote:
> >
> > (snip)
> >
> > > I don't know everything about IBM systems. You can stop a CPU on a VAX
> > > or Alpha from the console by issuing a control-p, which will produce
> > > the three chevrons >>>. The system will be stop. You can issue a
> > > "continue" and the system will continue were it stop at. There is a
> > > time limit and how much time you have I don't know.
> >
> > IBM S/360 and successors have a PSW bit called WAIT.
> >
> > When set, no instructions are executed, hopefully with
> > interrupts enabled. There is no NULL job or NULL task.
> >
> > The is especially useful when running on a virtual machine,
> > as VM doesn't have to use up 100% of its time running a NULL
> > task loop.
> >
> > The real reason is that IBM used to rent machines, such that the
> > rental charge was based on time actually spent processing.
> > A hardware timer could measure the amount actually used.
> >
> > As far as I know, DEC never rented VAX or Alpha that way.
> >
> > -- glen
>
> Dear Glen Herrmannsfeldt:
>
> I should have said: "I don't know anything about IBM".
>
> There were businesses that rented time out on a VAX/VMS. However, you
> can't charge by cpu runtime. You have to charge by connect time. This
> is due to the runtime not being the same each time the process is
> executed. The difference really showed itself when a process ran with
> no load on the VAX and then on a loaded VAX. This is due to the VMS
> scheduler.
>
Going back 20 years and more, that was always a problem for mainframers
coming to VMS. They had been used to charging usage out to other
departments according to CPU, I/O and disk usage, but VMS made it
difficult to pin CPU and I/O usage down the same way, especially when
using a single process to service requests for multiple users.
-- Paul Sture
- Next message: John Santos: "Re: Why was VAX abandonned ?"
- Previous message: JF Mezei: "Re: HP/Intel Itanium is dead, long life to HP/AMD Athlon"
- In reply to: Daryl Jones: "Re: Why was VAX abandonned ?"
- Next in thread: JF Mezei: "Re: Why was VAX abandonned ?"
- Reply: JF Mezei: "Re: Why was VAX abandonned ?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|