20 November  2007   8:51:28.351 AM      
 
COMPUTE_PI
  FORTRAN77 version
 
  Estimate PI by summing a series.
 
  This program includes Open MP directives
  to run the program in parallel.
 
  The number of processors available:
  OMP_GET_NUM_PROCS () =        1
 
  Call OMP_SET_NUM_THREADS, and request        4 threads.
  This is process   0
 
  Calling OMP_GET_NUM_THREADS inside a 
  parallel region, we get the number of
  threads is        1
 
 
R4_TEST:
  Estimate the value of PI,
  using single precision arithmetic.
 
  N = number of terms computed and added;
 
  MODE = SEQ for sequential code;
  MODE = OMP for Open MP enabled code;
  (performance depends on if Open MP is used,
  and how many processes are availablec)
 
  ESTIMATE = the computed estimate of PI;
 
  ERROR = ( the computed estimate - PI );
 
  TIME = elapsed wall clock time;
 
  You can't increase N forever, because:
  A) ROUNDOFF starts to be a problem, and
  B) maximum integer size is a problem.
 
           N Mode    Estimate        Error           Time
 
           1  SEQ    3.2000000477    0.584073E-01     0.00000    
           1  OMP    3.2000000477    0.584073E-01     0.00000    
          10  SEQ    3.1424260139    0.833273E-03     0.00000    
          10  OMP    3.1424260139    0.833273E-03     0.00000    
         100  SEQ    3.1416008472    0.810623E-05     0.00000    
         100  OMP    3.1416008472    0.810623E-05     0.00000    
        1000  SEQ    3.1415929794    0.238419E-06     0.00000    
        1000  OMP    3.1415929794    0.238419E-06     0.00000    
       10000  SEQ    3.1415905952    0.214577E-05    0.100000E-02
       10000  OMP    3.1415905952    0.214577E-05     0.00000    
      100000  SEQ    3.1415328979    0.598431E-04    0.600000E-02
      100000  OMP    3.1415328979    0.598431E-04    0.600000E-02
     1000000  SEQ    3.1413483620    0.244379E-03    0.580000E-01
     1000000  OMP    3.1413483620    0.244379E-03    0.580000E-01
 
R8_TEST:
  Estimate the value of PI,
  using double precision arithmetic.
 
  N = number of terms computed and added;
 
  MODE = SEQ for sequential code;
  MODE = OMP for Open MP enabled code;
  (performance depends on if Open MP is used,
  and how many processes are availablec)
 
  ESTIMATE = the computed estimate of PI;
 
  ERROR = ( the computed estimate - PI );
 
  TIME = elapsed wall clock time;
 
  You can't increase N forever, because:
  A) ROUNDOFF starts to be a problem, and
  B) maximum integer size is a problem.
 
             N Mode    Estimate        Error           Time
 
             1  SEQ    3.2000000000    0.584073E-01     0.00000    
             1  OMP    3.2000000000    0.584073E-01     0.00000    
            10  SEQ    3.1424259850    0.833331E-03     0.00000    
            10  OMP    3.1424259850    0.833331E-03     0.00000    
           100  SEQ    3.1416009869    0.833333E-05     0.00000    
           100  OMP    3.1416009869    0.833333E-05     0.00000    
          1000  SEQ    3.1415927369    0.833333E-07     0.00000    
          1000  OMP    3.1415927369    0.833333E-07     0.00000    
         10000  SEQ    3.1415926544    0.833341E-09    0.100000E-02
         10000  OMP    3.1415926544    0.833341E-09    0.100000E-02
        100000  SEQ    3.1415926536    0.836842E-11    0.700000E-02
        100000  OMP    3.1415926536    0.836842E-11    0.700000E-02
       1000000  SEQ    3.1415926536    0.284217E-13    0.710000E-01
       1000000  OMP    3.1415926536    0.284217E-13    0.710000E-01
      10000000  SEQ    3.1415926536    0.621725E-13    0.715000    
      10000000  OMP    3.1415926536    0.621725E-13    0.713000    
     100000000  SEQ    3.1415926536    0.633271E-12     7.15200    
     100000000  OMP    3.1415926536    0.633271E-12     7.13000    
    1000000000  SEQ    3.1415926536    0.177636E-12     71.5750    
    1000000000  OMP    3.1415926536    0.177636E-12     71.3210    
 
COMPUTE_PI
  Normal end of execution.
 
20 November  2007   8:54:07.248 AM