Ran a slightly larger data set size (yes, 300,000,000)...
I think I like these results, they are making more sense. Now
to see if smaller size make sense...
Brian
keeper@sae3:/home/keeper/stream% 208) setenv PARALLEL 64
keeper@sae3:/home/keeper/stream% 209) pstx2-64
----------------------------------------------
Double precision appears to have 16 digits of accuracy
Assuming 8 bytes per DOUBLE PRECISION word
----------------------------------------------
Array size = 300000000
Offset = 0
The total memory requirement is 6866 MB
You are running each test 10 times
The *best* time for each test is used
----------------------------------------------------
Your clock granularity appears to be less than one microsecond
Your clock granularity/precision appears to be 1 microseconds
The tests below will each take a time on the order
of 605053 microseconds
(= 605053 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: 12071.9246 0.4192 0.3976 0.4464
Scale: 7967.1486 0.6185 0.6025 0.6578
Add: 8497.6855 0.8515 0.8473 0.8592
Triad: 8590.0299 0.8652 0.8382 0.8755
Sum of a is = 3.4599023433963D+20
Sum of b is = 6.9198046533463D+19
Sum of c is = 9.2264062741574D+19
Note: Nonstandard floating-point mode enabled
See the Numerical Computation Guide, ieee_sun(3M)
This archive was generated by hypermail 2b29 : Tue Apr 18 2000 - 05:23:08 CDT