Automatic running test with large-scale systems

In some cases, one may want to know machine performance for more time consuming calculations. For this purpose, an automatic running test with relatively large-scale systems can be performed by

For the serial running

     % ./openmx -runtestL
  

For the MPI parallel running

     % mpirun -np 4 openmx -runtestL
  

For the OpenMP/MPI parallel running

     % mpirun -np 4 openmx -runtestL -nt 1
  

Then, OpenMX will run with 20 test files, and compare calculated results with the reference results which are stored in 'work/large_example'. The comparison (absolute difference in the total energy and force) is stored in a file 'runtestL.result' in the directory 'work'. The reference results were calculated using 40 MPI processes of a 2.6GHz Opteron cluster machine. If the difference is within last seven digits, we may consider that the installation is successful. As an example, 'runtestL.result' generated by the automatic running test is shown below:

1	large_example/5_5_13COb2.dat	Elapsed time(s)= 1740.27	diff Utot= 0.000000000038	diff Force= 0.000000000002
2	large_example/B2C62_Band.dat	Elapsed time(s)= 5035.37	diff Utot= 0.000000015973	diff Force= 0.000000006675
3	large_example/CG15c-Kry.dat	Elapsed time(s)= 1298.85	diff Utot= 0.000000001480	diff Force= 0.000000002069
4	large_example/DIA512-1.dat	Elapsed time(s)= 615.16	diff Utot= 0.000000033780	diff Force= 0.000000009994
5	large_example/FM.dat	Elapsed time(s)= 4533.59	diff Utot= 0.000000000018	diff Force= 0.000000000000
6	large_example/FeBCC.dat	Elapsed time(s)= 4128.44	diff Utot= 0.000000005436	diff Force= 0.000000000002
7	large_example/GEL.dat	Elapsed time(s)= 764.81	diff Utot= 0.000000000006	diff Force= 0.000000000000
8	large_example/GFRAG.dat	Elapsed time(s)= 801.09	diff Utot= 0.000000000002	diff Force= 0.000000000001
9	large_example/GGFF.dat	Elapsed time(s)=19139.17	diff Utot= 0.000000000026	diff Force= 0.000000000004
10	large_example/MCCN.dat	Elapsed time(s)= 2201.42	diff Utot= 0.000000000104	diff Force= 0.000000000065
11	large_example/Mn12_148_F.dat	Elapsed time(s)= 1538.40	diff Utot= 0.000000000014	diff Force= 0.000000000000
12	large_example/N1C999.dat	Elapsed time(s)= 3765.41	diff Utot= 0.000000013076	diff Force= 0.000000017373
13	large_example/Ni63-O64.dat	Elapsed time(s)= 2540.13	diff Utot= 0.000000000220	diff Force= 0.000000000126
14	large_example/Pt63.dat	Elapsed time(s)= 1718.62	diff Utot= 0.000000016353	diff Force= 0.000000000012
15	large_example/SialicAcid.dat	Elapsed time(s)= 204.23	diff Utot= 0.000000000278	diff Force= 0.000000000139
16	large_example/aAFM.dat	Elapsed time(s)= 5934.01	diff Utot= 0.000000000021	diff Force= 0.000000000001
17	large_example/cAFM.dat	Elapsed time(s)= 3994.10	diff Utot= 0.000000000022	diff Force= 0.000000000001
18	large_example/gAFM.dat	Elapsed time(s)= 5326.87	diff Utot= 0.000000000017	diff Force= 0.000000000001
19	large_example/nsV4Bz5.dat	Elapsed time(s)= 2675.28	diff Utot= 0.000000000394	diff Force= 0.000000000147
20	large_example/opt_4T2L_n.dat	Elapsed time(s)=27734.05	diff Utot= 0.000000000006	diff Force= 0.000000000005

Total elapsed time (s) 95689.26

The comparison was made using 20 processes by MPI and 2 threads by OpenMP (totally 40 cores) on the same machine. Since the automatic running test requires considerable memory size, you may encounter a segmentation fault on computational environment with small memory. Also the above example implies that the total elapsed time is more than 1 day even using 40 cores.