home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ogicse!das-news.harvard.edu!cantaloupe.srv.cs.cmu.edu!murthy
- From: murthy+@EUPHRATES.EDRC.CMU.EDU (Sesh Murthy)
- Newsgroups: comp.unix.aix,cmu.cs.unix.forum,cmu.cs.general
- Subject: Comparisons of running time. 13.25:1 difference!
- Message-ID: <1992Jul23.155257.107592@cs.cmu.edu>
- Date: 23 Jul 92 07:52:57 GMT
- Article-I.D.: cs.1992Jul23.155257.107592
- Organization: School of Computer Science, Carnegie Mellon
- Lines: 21
- Nntp-Posting-Host: euphrates.edrc.cmu.edu
- Originator: murthy@EUPHRATES.EDRC.CMU.EDU
-
- I am running my code on IBM 320s and DEC 5000s. You may find
- the ratio of running times interesting. Please tell me if I am
- comparing apples and oranges.
- Machine no iterations time in minutes flags for optimization
-
- Dec 5000/200 328328 iterations 580:26 cpu minutes -O
- IBM 320 5835830 iterations 778:19 cpu minutes -Q -O -qrrm
-
- The IBM 320 was 13.25 times faster than the DEC 5000/200!
-
- So is this MACH lossage or what? If you are running any
- compute intensive tasks take note.
-
- Sesh Murthy
-
- Notes: -O3 on the DEC machine did not terminate after 800 minutes and
- was not used. The Dec machine (gaia) was running MACH 2.5.
- The process is compute bound. The code is in C and has a fair amount of
- pointer chasing and floating point operations. The code on the two
- machines was identical. I used the XLC compiler on the IBM and the
- cc compiler on the 5000.
-