Convex SPP-1200 120 Mhz HP 7200 chip. This is the result of 1 cpu.
fc -O2 ( the output of compiling said the 4 loops (copy,scale,sum,triad)
are unrolled 6, 6, 5, 5 times, respectively )
----------------------------------------------
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
----------------------------------------------------
Your clock granularity/precision appears to be 9999 microseconds
The tests below will each take a time on the order
of 600000 microseconds
(= 60 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: 65.3059 .6430 .4900 .8500
Scaling : 62.7453 .8065 .5100 .8800
Summing : 70.5882 .8485 .6800 1.0000
SAXPYing : 72.7273 .8687 .6600 1.0300
Sum of a is : 2.306601562566441E+18
Sum of b is : 4.613203125032299E+17
Sum of c is : 6.150937500126260E+17
----------------------------------------------
Convex SPP-1200 120 Mhz HP 7200 chip. This is the result of 1 cpu.
fc -O2 -nur ( nur means no unroll )
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
----------------------------------------------------
Your clock granularity/precision appears to be 9999 microseconds
The tests below will each take a time on the order
of 600000 microseconds
(= 60 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: 54.2373 .8079 .5900 .9300
Scaling : 30.1887 1.1813 1.0600 1.4000
Summing : 71.6420 .9198 .6700 1.0900
SAXPYing : 47.0588 1.2820 1.0200 1.3600
Sum of a is : 2.306601562591873E+18
Sum of b is : 4.613203124856438E+17
Sum of c is : 6.150937500141255E+17
convex spp-120 120 Mhz HP 7200 chip. 8 cpus.
fc -O3 ( -O3 turns on parallel, but I don't know how many cpus it is
using. I am told by the consultant at computer center it is
using all cpus, which is 8 )
----------------------------------------------
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
----------------------------------------------------
Your clock granularity/precision appears to be 9998 microseconds
The tests below will each take a time on the order
of 1430002 microseconds
(= 143 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: 35.1647 1.2518 .9100 1.7300
Scaling : 35.1650 1.2726 .9100 1.5300
Summing : 40.3363 1.3175 1.1900 1.6500
SAXPYing : 46.6018 1.2807 1.0300 1.5600
Sum of a is : 2.306601562496450E+18
Sum of b is : 4.613203124994498E+17
Sum of c is : 6.150937500002008E+17
This archive was generated by hypermail 2b29 : Tue Apr 18 2000 - 05:23:05 CDT