home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!sdd.hp.com!cs.utexas.edu!sun-barr!male.EBay.Sun.COM!exodus.Eng.Sun.COM!rbbb.Eng.Sun.COM!chased
- From: chased@rbbb.Eng.Sun.COM (David Chase)
- Newsgroups: comp.arch
- Subject: Re: Sun 600MP Benchmark Anomaly
- Date: 21 Jul 1992 17:25:50 GMT
- Organization: Sun Microsystems, Mt. View, Ca.
- Lines: 24
- Message-ID: <l6oi4uINN62e@exodus.Eng.Sun.COM>
- References: <5623@nosc.NOSC.MIL> <19190001@hpcss01.cup.hp.com>
- NNTP-Posting-Host: rbbb
-
- In article <19190001@hpcss01.cup.hp.com> markw@hpcss01.cup.hp.com (Mark Williams) writes:
- >But the patch didn't really fix the anomaly. It helped two processor
- >performance, but did not change the fact that 4-processor system ran no
- >faster than a two processor system, at least in all the benchmarks I've
- >seen.
-
- On the only benchmark that matters to me, they (running 4.1.2,
- probably with the patch but I really don't know) work great. Of
- course, my benchmark suite is (optimizing) compiler testing. If we
- had twice as many 4P 600s, I'd be twice as happy, because I'd test
- twice as much.
-
- >Bottom line: the 600MP (with SunOS 4 and the May patch) performs poorly as
- >an MP system, but it is a fine uniprocessor with better I/O than a workstation.
-
- In my (limited) experience it is a fine MP system. Careful with those
- blatant generalizations, sir. I think I could make a few myself
- about, say, the use{ful,less}ness of pattern-matching compiler
- optimizations on non-SPEC codes.
-
- Bottom line: Beware of benchmarks.
-
- David Chase
- Sun
-