new STREAM results

From: Mikhail Kuzminsky (kus@free.net)
Date: Mon Sep 02 2002 - 13:57:58 CDT

  • Next message: Craig Armour: "G4 Stream results (partially experimental)"

        Dear Prof. McCalpin,
    I'm sending you results of STREAM benchmark run for
    5 different computers(w/different CPUs and chipsets and memory
    type) performed by me in our computer center (on the nodes of
    our clusters) inclusion in "official" table of results.

    Yours
    Dr. Mikhail Kuzminsky
    Zelinsky Institute of Organic Chemsitry RAS
    Computer Assistance to Chemical Research Center
    Moscow
    ==============================================

      The conditions common for all the tests
      a) All the tests were performed w/using of ifc 5.0 compiler
         (compilation w/options -tpp6 -O3)
       b) All the tests were performed w/using of high resolution
          RDTSC-based timer
       c) All the tests were performed w/same size of arrays
          (2000000)

    1) ASUS P2B-F, chipset 440BX, memory ECC PC100
       Intel Celeron/433 Mhz (cache L2=128 Kbytes)

    ----------------------------------------------
     Double precision appears to have 16 digits of accuracy
     Assuming 8 bytes per DOUBLE PRECISION word
    ----------------------------------------------
     Array size = 2000000
     Offset = 0
     The total memory requirement is 45 MB
     You are running each test 10 times
     --
     The *best* time for each test is used
     *EXCLUDING* the first and last iterations
     ----------------------------------------------------
     Your clock granularity appears to be less than one microsecond
     Your clock granularity/precision appears to be 1 microseconds
     ----------------------------------------------------
    Function Rate (MB/s) Avg time Min time Max time
    Copy: 203.7948 0.1571 0.1570 0.1571
    Scale: 204.8529 0.1563 0.1562 0.1563
    Add: 246.2016 0.1950 0.1950 0.1951
    Triad: 247.4185 0.1940 0.1940 0.1941
     ----------------------------------------------------
     Solution Validates!
     ----------------------------------------------------

    2) ASUS P2B-D /chipset i440BX, memory ECC PC100
       Intel Pentium III/600 Mhz cache L2=512 KBytes

    ----------------------------------------------
     Double precision appears to have 16 digits of accuracy
     Assuming 8 bytes per DOUBLE PRECISION word
    ----------------------------------------------
     Array size = 2000000
     Offset = 0
     The total memory requirement is 45 MB
     You are running each test 10 times
     --
     The *best* time for each test is used
     *EXCLUDING* the first and last iterations
     ----------------------------------------------------
     Your clock granularity appears to be less than one microsecond
     Your clock granularity/precision appears to be 1 microseconds
     ----------------------------------------------------
    Function Rate (MB/s) Avg time Min time Max time
    Copy: 311.0624 0.1029 0.1029 0.1029
    Scale: 313.8063 0.1020 0.1020 0.1021
    Add: 368.3702 0.1303 0.1303 0.1304
    Triad: 363.5981 0.1321 0.1320 0.1322
     ----------------------------------------------------
     Solution Validates!
     ----------------------------------------------------

    3) Gigabyte GA-7VX, chipset VIA KX133, memory ECC PC133
       AMD Athlon /700 Mhz, cache L2=512 Kbytes

    ----------------------------------------------
     Double precision appears to have 16 digits of accuracy
     Assuming 8 bytes per DOUBLE PRECISION word
    ----------------------------------------------
     Array size = 2000000
     Offset = 0
     The total memory requirement is 45 MB
     You are running each test 10 times
     --
     The *best* time for each test is used
     *EXCLUDING* the first and last iterations
     ----------------------------------------------------
     Your clock granularity appears to be less than one microsecond
     Your clock granularity/precision appears to be 1 microseconds
     ----------------------------------------------------
    Function Rate (MB/s) Avg time Min time Max time
    Copy: 511.6148 0.0626 0.0625 0.0626
    Scale: 503.1171 0.0636 0.0636 0.0637
    Add: 584.8932 0.0821 0.0821 0.0821
    Triad: 484.1748 0.0992 0.0991 0.0992
     ----------------------------------------------------
     Solution Validates!
     ----------------------------------------------------

    4) Tyan S2518 chipset ServerWorks ServerSet III LE3
       Intel Pentium III Tualatin /1266 Mhz

    ----------------------------------------------
     Double precision appears to have 16 digits of accuracy
     Assuming 8 bytes per DOUBLE PRECISION word
    ----------------------------------------------
     Array size = 2000000
     Offset = 0
     The total memory requirement is 45 MB
     You are running each test 10 times
     --
     The *best* time for each test is used
     *EXCLUDING* the first and last iterations
     ----------------------------------------------------
     Your clock granularity appears to be less than one microsecond
     Your clock granularity/precision appears to be 1 microseconds
     ----------------------------------------------------
    Function Rate (MB/s) Avg time Min time Max time
    Copy: 385.8379 0.0830 0.0829 0.0830
    Scale: 386.3428 0.0828 0.0828 0.0829
    Add: 515.7584 0.0937 0.0931 0.0938
    Triad: 510.4224 0.0941 0.0940 0.0945
     ----------------------------------------------------
     Solution Validates!
     ----------------------------------------------------
    5) Tyan S2460, chipset AMD760 MP, memory ECC DDR266
      Athlon XP/1800+ 1533 Mhz

    ----------------------------------------------
     Double precision appears to have 16 digits of accuracy
     Assuming 8 bytes per DOUBLE PRECISION word
    ----------------------------------------------
     Array size = 2000000
     Offset = 0
     The total memory requirement is 45 MB
     You are running each test 10 times
     --
     The *best* time for each test is used
     *EXCLUDING* the first and last iterations
     ----------------------------------------------------
     Your clock granularity appears to be less than one microsecond
     Your clock granularity/precision appears to be 1 microseconds
     ----------------------------------------------------
    Function Rate (MB/s) Avg time Min time Max time
    Copy: 787.4216 0.0407 0.0406 0.0407
    Scale: 771.5594 0.0415 0.0415 0.0415
    Add: 847.9618 0.0566 0.0566 0.0567
    Triad: 730.7458 0.0660 0.0657 0.0662
     ----------------------------------------------------
     Solution Validates!
     ----------------------------------------------------

    =======END OF DATA PRESENTED========



    This archive was generated by hypermail 2.1.4 : Fri Nov 08 2002 - 13:37:15 CST