Re: STREAM Bandwidth results

From: Michael Kukat (
Date: Wed Nov 13 2002 - 12:20:42 CST

  • Next message: Michael Kukat: "Re: STREAM Bandwidth results"

    Okay, me again...

    another rare box four your collection:

    Vendor: Tandem
    System: Integrity S2, MIPS R3000, i think 16 MHz, 48 MB RAM
    OS : Nonstop-UX 4.0
    Build : cc -o /tmp/stream -O3 -s stream_d.c second_wall.c -lm

    (McCalpin's Note: this box was first available in January 1990)

    Results stay constant over sevaral runs. I had to decrease the array size due
    to lack of enough RAM, but i increased the number of runs. But i think, this
    doesn't matter.


    This system uses 8 bytes per DOUBLE PRECISION word.
    Array size = 500000, Offset = 0
    Total memory required = 11.4 MB.
    Each test is run 50 times, but only
    the *best* time for each is used.
    Your clock granularity/precision appears to be 144 microseconds.
    Each test below will take on the order of 381381 microseconds.
       (= 2648 clock ticks)
    Increase the size of the arrays if this shows that
    you are not getting at least 20 clock ticks per test.
    WARNING -- The above is only a rough guideline.
    For best results, please be sure you know the
    precision of your system timer.
    Function Rate (MB/s) RMS time Min time Max time
    Copy: 22.5102 0.3644 0.3554 0.4315
    Scale: 21.5820 0.3790 0.3707 0.4050
    Add: 27.6243 0.4436 0.4344 0.4960
    Triad: 26.9698 0.4570 0.4449 0.5242


    --     Home Powered by: (Net|Open|Free)BSD IRIX NonStop-UX
    Solaris AIX HP-UX Tru64 MUNIX Ultrix VMS SINIX Dolphin_Unix OpenStep MacOS A/UX

    This archive was generated by hypermail 2.1.4 : Wed Jan 08 2003 - 09:17:51 CST