Re: Attainable memory bandwidth

From: Gary Tarolli (tarolli@tenno.boston.sgi.com)
Date: Thu Oct 03 1991 - 11:30:17 CDT


I reran the tests from Dave Olson's machines and got the following results
on my Indigo:

the following results were obtained by compiling on my PI running 3.3:

--------------------------------------
 Double precision appears to have 16 digits of accuracy
 Assuming 8 bytes per DOUBLEPRECISION word
--------------------------------------
 Timing calibration ; time = 52.00000219047070 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: 36.9233 0.1321 0.1300 0.1400
Scaling : 34.2857 0.1481 0.1400 0.1600
Summing : 36.0000 0.2051 0.2000 0.2100
SAXPYing : 36.0000 0.2111 0.2000 0.2200

the following results were obtained by compiling on my Indigo running 4.0

--------------------------------------
 Double precision appears to have 16 digits of accuracy
 Assuming 8 bytes per DOUBLEPRECISION word
--------------------------------------
 Timing calibration ; time = 51.99999623000622 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: 36.9231 0.1310 0.1300 0.1400
Scaling : 26.6667 0.1901 0.1800 0.2000
Summing : 32.7273 0.2230 0.2200 0.2300
SAXPYing : 30.0000 0.2480 0.2400 0.2500

Obviously there's a bug in the 4.0 compiler that produces suboptimal code.
However, even given that, I got better numbers than Dave. I am going to
submit a bug reporting that the performance of the code produced by the 4.0
compiler went down hill. I have look at the executable, and the 3.3 simply
does better floating point instruction scheduling.
______________________________________________________________________________
  _____ ______ _ _ (508)562-4800 tarolli@sgi.com
 / ___ __ __ / __ __ ___ // // * M/S DER-200
(____/ (_/_/ (_(_/ / (_/_/ (_(_/ (/_(/_/_
               _/



This archive was generated by hypermail 2b29 : Tue Apr 18 2000 - 05:23:01 CDT