| 32 | |
| 33 | == Benchmarks == |
| 34 | |
| 35 | Once you build the library and associated programs (using make), you can run, for example: |
| 36 | |
| 37 | {{{ |
| 38 | $ mpirun -np 8 ./mpi/GenericIOBenchmarkWrite /tmp/out.gio 123456 2 |
| 39 | Wrote 9 variables to /tmp/out (4691036 bytes) in 0.2361s: 18.9484 MB/s |
| 40 | }}} |
| 41 | |
| 42 | {{{ |
| 43 | $ mpirun -np 8 ./mpi/GenericIOBenchmarkRead /tmp/out.gio |
| 44 | Read 9 variables from /tmp/out (4688028 bytes) in 0.223067s: 20.0426 MB/s [excluding header read] |
| 45 | }}} |
| 46 | |
| 47 | The read benchmark always reads all of the input data. The output benchmark takes two numerical parameters, one if the number of data rows to write, and the second is a random seed (which slightly perturbs the per-rank output sizes, but not by much). Each row is 36 bytes for these benchmarks. |