Re : Benchmark on I/O calls

Article View: comp.sys.transputer

Article #830

Back to comp.sys.transputer Previous Next

#830

From: YEEJJ
Date: Mon, 30 Jul 1990 20:31

29 lines
1413 bytes


        I have an Inmos B008 with a T800-20 as the root transputer. The host
 is a 386/25 with SCSI disk. I compiled the program by Vandana on the 3L
 parallel C. Since the PC has a turbo switch (20Mhz/25Mhz), I ran the test
 several times on each mode and used a software switch to change the memory
 between zeo and one wait state. Below are the results :

                0 wait state            1 wait state

        25 Mhz     7.90 s                  8.81 s

        20 Mhz     9.22 s                 10.28 s

        The results are much worse than what Ken Koontz reported for his B008
 with a B404-3 root and on Logical Systems C. Is the problem with 3L parallel
 C, or with SCSI disk?

        Hoping to hear results from others. I would like to know people with
 experiences on 3L C too.

 Jenn-Jong, Yee
 National University of Singapore
 email : YeeJJ@nusdiscs


> YEE JEUU JONG (YEEJJ@NUSDISCS.BITNET)
> National University of Singapore
> Department of Information Systems and Computer Science

Technical Details

Message-ID: <9007300428.AA05037@tcgould.TN.CORNELL.EDU>

Path:

arc1-ca.pugleaf.net!utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!mips!decwrl!ucbvax!NUSDISCS.BITNET!YEEJJ

Show Full Headers

Complete Article Headers

Path: arc1-ca.pugleaf.net!utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!mips!decwrl!ucbvax!NUSDISCS.BITNET!YEEJJ
From: YEEJJ@NUSDISCS.BITNET
Newsgroups: comp.sys.transputer
Subject: Re : Benchmark on I/O calls
Message-ID: <9007300428.AA05037@tcgould.TN.CORNELL.EDU>
Date: 30 Jul 90 20:31:00 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 29