30 March 2023   8:30:35.664 AM
 
RANDOM_NUMBER_CONTENTION():
  FORTRAN90 version
 
  This program demonstrates a case where the use
  of multiple threads in OpenMP may DEGRADE the
  performance of a program.
 
  In this program, we simply call the FORTRAN90
  random number generator many times.
 
  Because that function has internal saved variables,
  when it is accessed by multile threads, there can
  be huge delays because the threads fight to gain
  access to the saved variables.
 
  The number of processors available:
  OMP_GET_NUM_PROCS () =       16
 
   Threads    Wallclock time
              (seconds)
 
         1    0.171848E-01
         2    0.106330E-01
         3    0.843619E-02
         4    0.545172E-02
         5    0.542310E-02
         6    0.387512E-02
         7    0.353254E-02
         8    0.462323E-02
 
RANDOM_NUMBER_CONTENTION():
  Normal end of execution.
 
30 March 2023   8:30:35.723 AM
30 March 2023   8:30:35.726 AM
 
RANDOM_NUMBER_CONTENTION():
  FORTRAN90 version
 
  This program demonstrates a case where the use
  of multiple threads in OpenMP may DEGRADE the
  performance of a program.
 
  In this program, we simply call the FORTRAN90
  random number generator many times.
 
  Because that function has internal saved variables,
  when it is accessed by multile threads, there can
  be huge delays because the threads fight to gain
  access to the saved variables.
 
  The number of processors available:
  OMP_GET_NUM_PROCS () =       16
 
   Threads    Wallclock time
              (seconds)
 
         1    0.192689E-01
         2    0.832213E-02
         3    0.573141E-02
         4    0.411774E-02
         5    0.454677E-02
         6    0.412201E-02
         7    0.358064E-02
         8    0.409353E-02
 
RANDOM_NUMBER_CONTENTION():
  Normal end of execution.
 
30 March 2023   8:30:35.779 AM
30 March 2023   8:30:35.781 AM
 
RANDOM_NUMBER_CONTENTION():
  FORTRAN90 version
 
  This program demonstrates a case where the use
  of multiple threads in OpenMP may DEGRADE the
  performance of a program.
 
  In this program, we simply call the FORTRAN90
  random number generator many times.
 
  Because that function has internal saved variables,
  when it is accessed by multile threads, there can
  be huge delays because the threads fight to gain
  access to the saved variables.
 
  The number of processors available:
  OMP_GET_NUM_PROCS () =       16
 
   Threads    Wallclock time
              (seconds)
 
         1    0.165208E-01
         2    0.826012E-02
         3    0.561571E-02
         4    0.422362E-02
         5    0.415974E-02
         6    0.326929E-02
         7    0.383789E-02
         8    0.267833E-02
 
RANDOM_NUMBER_CONTENTION():
  Normal end of execution.
 
30 March 2023   8:30:35.830 AM
30 March 2023   8:30:35.831 AM
 
RANDOM_NUMBER_CONTENTION():
  FORTRAN90 version
 
  This program demonstrates a case where the use
  of multiple threads in OpenMP may DEGRADE the
  performance of a program.
 
  In this program, we simply call the FORTRAN90
  random number generator many times.
 
  Because that function has internal saved variables,
  when it is accessed by multile threads, there can
  be huge delays because the threads fight to gain
  access to the saved variables.
 
  The number of processors available:
  OMP_GET_NUM_PROCS () =       16
 
   Threads    Wallclock time
              (seconds)
 
         1    0.164042E-01
         2    0.839801E-02
         3    0.558614E-02
         4    0.436359E-02
         5    0.439140E-02
         6    0.340850E-02
         7    0.384521E-02
         8    0.422902E-02
 
RANDOM_NUMBER_CONTENTION():
  Normal end of execution.
 
30 March 2023   8:30:35.882 AM