1 Some explanations on the MPI implementation of a subset of NPB 3.3 (NPB3.3-MPI)
2 ----------------------------------------------------------------------
4 NPB-MPI is a sample MPI implementation based on NPB2.4 and NPB3.0-SER.
5 This subset contains three of the original benchmarks: one in Fortran: EP;
6 one in C: IS, as well as the DT benchmark, written in C, introduced in NPB3.2-MPI.
10 Before compilation, one needs to check the configuration file
11 'make.def' in the config directory and modify the file if necessary.
13 make <benchmark-name> NPROCS=<number> CLASS=<class> \
14 [SUBTYPE=<type>] [VERSION=VEC]
16 where <benchmark-name> is "dt", "ep", or "is",
17 <number> is the number of processes
18 <class> is "S", "W", "A", "B", "C", "D", or "E"
20 Class E is not available for IS and DT.
22 Class D for IS (Integer Sort) requires a compiler/system that
23 supports the "long" type in C to be 64-bit.
27 The executable is named <benchmark-name>.<class>.<nprocs>
28 The executable is placed in the bin subdirectory (or in the directory
29 BINDIR specified in make.def, if you've defined it). The method for
30 running the MPI program depends on your local system.