Next: 19990902:
Up: MD work on
Previous: 19990825: with a different
19990831:
To be compared with 19990824 (41)/19990825 (42) to see
the effect of zirconia addition on the oxygen diffusion coefficient.
From now on I'm using the mpi_bsend version of
DL_POLY, in an attempt to solve the ``deadlock'' problem. The
mpi_bsend version means that I have recompiled DL_POLY
after substituting all the mpi_send MPI calls for
corresponding mpi_bsend ones, which should hopefully be safer
with respect to possible deadlock problems (following a suggestion of
CINECA consultant dr. Sigismondo Boschi)