home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cs.utexas.edu!sun-barr!ames!tulane!ukma!cs.widener.edu!dsinc!ub!niktow!pavlov
- From: pavlov@niktow.canisius.edu (Greg Pavlov)
- Newsgroups: comp.sys.sun.hardware
- Subject: Re: Is Sparc out of Gas?
- Message-ID: <1898@niktow.canisius.edu>
- Date: 18 Dec 92 08:02:51 GMT
- References: <lg6j1iINN5dj@exodus.Eng.Sun.COM> <jdd.723233663@cdf.toronto.edu> <1992Dec15.081616.14308@irfu.se>
- Organization: Canisius College, Buffalo NY. 14208
- Lines: 34
-
- In article <1992Dec15.081616.14308@irfu.se>, bt@irfu.se (Bo Thide') writes:
- > In article <1784@niktow.canisius.edu> pavlov@niktow.canisius.edu (Greg Pavlov) writes:
- > >In article <jdd.723233663@cdf.toronto.edu>, jdd@cdf.toronto.edu (John DiMarco) writes:
- > >> In <1767@niktow.canisius.edu> pavlov@niktow.canisius.edu (Greg Pavlov) writes:
- > >>
- > >> Everybody's going to have to do MP eventually. Sun chose to do it now rather
- > >> than later.
- > >>
- > > The other vendors decided to defer ??? And "... do MP eventually" ??? A
- > > lot of VAX 6nnn, DECsystem 58nn, SGI ____ users, to name a few, would be
- > > surprised to hear that...
- >
- > So would HP, who had MP alrady in their HP9000/500 UNIX boxes of 1983/84
- > vintage! Interestingly enough, the underlying OS on the 500 series was called
- > SUN!
- >
- Well, I seriously considered mentioning the 9000/500 AND its vintage in my
- note: that particular machine, with two processors, was the first UNIX box
- I purchased (my company's money and thus box, actually). We used it for apx.
- two years but when I saw the handwriting on the wall - HP was obviously not
- going to continue with it - I sold it off and moved on to something else.
-
- The reason I decided NOT to mention it is that while yes, it was a MP system,
- I am not sure what that meant in that particular system: if one ran a user-
- level application, one could see the two processors "take turns" running it
- for a second at a time, e.g., it seemed that the mp mechanism was either
- rather slow and/or the relative overhead was so high that relatively long
- time slices were used. Much of that may have been due to the low throughput
- of the processors - each generally on a par with the VAX 730. For its time,
- though, the 9000/500 was a very nice system.
-
-
- greg pavlov
- pavlov@fstrf.org
-