User Tools

Site Tools

Table of Contents


The NAS Parallel Benchmark is a set of multiple programs, both pseudo application benchmarks and synthetic benchmark, to benchmark different aspects of parallel supercomputers, among them is the BT-IO benchmark designed specifically to test different approaches to parallel I/O.



Create a make.def file in the config directory from the given template and make sure to change all the include and library path and the compilers to the correct ones. Compile the BT-I/O benchmark using make bt CLASS=B NPROCS=1 type=full where NPROCS is the number of processes to be used when executing the binary.

Example Output

Here is an example run of bt-io with a class A workload and 4 processes:

mpirun -np 4 bt.A.4

 NAS Parallel Benchmarks 3.3 -- BT Benchmark 

 No input file Using compiled defaults
 Size:   64x  64x  64
 Iterations:  200    dt:   0.0008000
 Number of active processes:     4

 Time step    1
 Time step   20
 Time step   40
 Time step   60
 Time step   80
 Time step  100
 Time step  120
 Time step  140
 Time step  160
 Time step  180
 Time step  200
 Verification being performed for class A
 accuracy setting for epsilon =  0.1000000000000E-07
 Comparison of RMS-norms of residual
           1 0.1080634671464E+03 0.1080634671464E+03 0.7101254243319E-14
           2 0.1131973090122E+02 0.1131973090122E+02 0.9415542762816E-15
           3 0.2597435451158E+02 0.2597435451158E+02 0.6838887328685E-15
           4 0.2366562254468E+02 0.2366562254468E+02 0.6455215276928E-14
           5 0.2527896321175E+03 0.2527896321175E+03 0.1259241302557E-13
 Comparison of RMS-norms of solution error
           1 0.4234841604053E+01 0.4234841604053E+01 0.1048656009767E-14
           2 0.4439028249700E+00 0.4439028249700E+00 0.6752834174215E-14
           3 0.9669248013635E+00 0.9669248013635E+00 0.3100139909806E-14
           4 0.8830206303977E+00 0.8830206303977E+00 0.6286506715733E-15
           5 0.9737990177083E+01 0.9737990177083E+01 0.1824151397873E-14
 Verification Successful

 BT Benchmark Completed.
 Class           =                        A
 Size            =             64x  64x  64
 Iterations      =                      200
 Time in seconds =                    46.24
 Total processes =                        4
 Compiled procs  =                        4
 Mop/s total     =                  3639.24
 Mop/s/process   =                   909.81
 Operation type  =           floating point
 Verification    =               SUCCESSFUL
 Version         =                    3.3.1
 Compile date    =              31 May 2017

 Compile options:
    MPIF77       = mpif77
    FLINK        = $(MPIF77)
    FMPI_LIB     = -L/usr/lib64 -lmpi
    FMPI_INC     = -I/usr/include
    FFLAGS       = -O
    RAND         = (none)

 Please send feedbacks and/or the results of this run to:

 NPB Development Team