On Tue, Oct 17, 2006, Ross Walker wrote:
>
> A quick test and it seems that sander.PIMD is not being built by default
> either in serial or parallel but the default test suite is setup to run the
> PIMD test cases.
The sander.PIMD executable no longer exists. It does look like the Makefile
in amber10/test needs to be updated. Note that PIMD runs (as I understand it)
now require a parallel build (i.e. use sander.MPI).
Users should note (I hope I am correct here!) that PIMD runs are now set up in
a new way: you use a "classical" (ordinary) prmtop file, and do not need to
process it through addles. The old test cases in amber10/test/pimd_* look to
be obsolete; looks like the new examples are in amber10/test/full_pimd.
Several of us will be working on cleaning up the test directory, but Wei
should take the lead on this.
In addition, presumably(?) the combination of EVB + PIMD does not now work.
EVB + classical MD/minimization should still be OK.
>
> On another note what is the status of NEB under the new system? Does it need
> to be run with sander.PIMD and I assume sander.PIMD.MPI? Also are things
> linear in memory for this as well?
The idea is that NEB will work under sander.MPI, with a linear memory layout
analogous (identical?) to that used for PIMD.
>
> On one other note ;-) does this mean a lot the ifdef PIMD stuff will be
> going away?
Yes.
> > simulation type, "ipimd=0" means normal PIMD simulation,
^^^^^^^^^^^
normal (non-PIMD)
...dac
Received on Wed Oct 18 2006 - 06:07:30 PDT
