New results: 56-cpu 8400

From: John, dtn 381-0378 21-Jan-1997 1122 (henning@perfom.ENET.dec.com)
Date: Tue Jan 21 1997 - 10:29:42 CST


Hi John,

Here's a 56-CPU AlphaServer 8400 5/440 cluster (7 nodes).
The program was the same as what I submitted to you for
the 8400 5/350 (4-node, 32 cpu) except for a bit larger
array size.

How's life at SGI?

Best regards
        /john

% ./56cpu
----------------------------------------------
 Double precision has 15 digits of accuracy
 Assuming 8 bytes per DOUBLE PRECISION word
----------------------------------------------
 Array size = 60000000
 Offset = 0
 The total memory requirement is 1373 MB
 You are running each test 10 times
 The *best* time for each test is used
 ----------------------------------------------------
 Your clock granularity/precision appears to be 2900 microseconds
 The tests below will each take a time on the order
 of 132800 microseconds
    (= 46 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
Assignment: 7811.2286 0.1449 0.1229 0.1797
Scaling : 5617.3201 0.1850 0.1709 0.1814
Summing : 8328.5136 0.1966 0.1729 0.2585
SAXPYing : 6161.7458 0.2640 0.2337 0.3551

 Sum of a is : 6.919804687693033E+019
 Sum of b is : 1.383960937522879E+019
 Sum of c is : 1.845281250037503E+019



This archive was generated by hypermail 2b29 : Tue Apr 18 2000 - 05:23:06 CDT