home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!sun-barr!ames!saimiri.primate.wisc.edu!zaphod.mps.ohio-state.edu!ub!niktow!pavlov
- From: pavlov@niktow.canisius.edu (Greg Pavlov)
- Newsgroups: comp.sys.sun.hardware
- Subject: Re: How does the new 2000+MIP Sun server compare against a Cray/XMP?
- Message-ID: <1729@niktow.canisius.edu>
- Date: 18 Nov 92 21:52:19 GMT
- References: <lg2hufINNj7k@appserv.Eng.Sun.COM> <1992Nov11.185638.6416@PacBell.COM> <1e9bpoINNnpl@armory.centerline.com>
- Organization: Canisius College, Buffalo NY. 14208
- Lines: 20
-
- In article <1e9bpoINNnpl@armory.centerline.com>, matt@centerline.com (Matt Landau) writes:
- > Okay, it's time to set the record straight.
- >
- > Well, I've now heard from several people inside Sun, and although the
- > reports are slightly varied, the consensus seems to be that 20-CPU models
- > are, in fact, running inside Sun, and that the delay in shipping is more
- > to assure reliability and tune the OS than to actually make the basic
- > hardware work.
- >
- I'm sure that there is no question that 20-cpu systems are, in fact, running
- within the bowels of SUN's tech offices. But whether the total throughput
- of these systems is anywhere near 20*(throughput of a one-processor system)
- is suspect. Maybe that's what all of this "OS tuning" is all about. Then
- again, I would love to be proved wrong by some independently verified pub-
- lished "SPECrates"....
-
-
-
- greg pavlov
- pavlov@fstrf.org
-