STREAM

From: Randy Zagar (zagar@chester)
Date: Mon Apr 03 1995 - 16:49:49 CDT


John,

I did the following STREAM benchmarks on three Sun machines.
I had to decrease the value of 'n' to 1,000,000 so that it would
run on Clark (only has 48-MB RAM). I used the standard SUNPRO
compilers with -fast -O3 options.

How does machine load affect these results, John ?

 -Randy

#############################################################################

Strauss: (SS-2000)
--------------------------------------
 Double precision appears to have 16 digits of accuracy
 Assuming 8 bytes per DOUBLEPRECISION word
--------------------------------------
 Timing calibration ; time = 227.80580222607 hundredths of a second
 Increase the size of the arrays if this is <30
  and your clock precision is =<1/100 second
 ---------------------------------------------------
Function Rate (MB/s) RMS time Min time Max time
Assignment: 34.2462 0.4783 0.4672 0.4954
Scaling : 32.7680 0.5158 0.4883 0.6406
Summing : 34.9483 0.7315 0.6867 0.8021
SAXPYing : 34.9022 0.7168 0.6876 0.8027
 Sum of a is : 5.7665039064002D+17
 Sum of b is : 1.1533007812683D+17
 Sum of c is : 1.5377343750306D+17

Chester: (SS1)
--------------------------------------
 Double precision appears to have 16 digits of accuracy
 Assuming 8 bytes per DOUBLEPRECISION word
--------------------------------------
 Timing calibration ; time = 804.82341051102 hundredths of a second
 Increase the size of the arrays if this is <30
  and your clock precision is =<1/100 second
 ---------------------------------------------------
Function Rate (MB/s) RMS time Min time Max time
Assignment: 11.7287 1.3848 1.3642 1.4157
Scaling : 9.8648 1.6469 1.6219 1.6887
Summing : 12.0624 2.0166 1.9896 2.0492
SAXPYing : 10.4403 2.3244 2.2988 2.3465
 Sum of a is : 5.7665039064002D+17
 Sum of b is : 1.1533007812683D+17
 Sum of c is : 1.5377343750306D+17

Clark: ('classic)
--------------------------------------
 Double precision appears to have 16 digits of accuracy
 Assuming 8 bytes per DOUBLEPRECISION word
--------------------------------------
 Timing calibration ; time = 411.12433224916 hundredths of a second
 Increase the size of the arrays if this is <30
  and your clock precision is =<1/100 second
 ---------------------------------------------------
Function Rate (MB/s) RMS time Min time Max time
Assignment: 44.5260 0.3598 0.3593 0.3610
Scaling : 30.9918 0.5169 0.5163 0.5177
Summing : 41.7150 0.5759 0.5753 0.5764
SAXPYing : 31.9179 0.7526 0.7519 0.7538
 Sum of a is : 5.7665039064002D+17
 Sum of b is : 1.1533007812683D+17
 Sum of c is : 1.5377343750306D+17



This archive was generated by hypermail 2b29 : Tue Apr 18 2000 - 05:23:04 CDT