search

numat / RASPA2

NO LONGER UPDATED. Use the official repository.
https://github.com/iraspa/RASPA2
Other
73 stars 72 forks source link

Update cbmc.c #13

Closed vargaslo closed 9 years ago

vargaslo commented 9 years ago

Update energy to include all urey-bradley instances

vargaslo commented 9 years ago

Hi Patrick,

I noticed that you have recently made updates to ewald.c in the Raspa code to address memory leaks. I have come across an issue that I wonder if you may be familiar with as it pertains to that part of the code.

When I attempt to make a rather large system (80Ax80Ax80A) by increasing UnitCells, I get a segfault. The core dump points to PrecomputeFixedEwaldContributions() at ewald.c

Is this something that could be remedied, or is it an inherent limitation for system size?

I appreciate any insight you may have on this.

Best, Ernesto

On Mon, Sep 15, 2014 at 1:11 PM, Patrick Fuller notifications@github.com wrote:

Merged #13 https://github.com/numat/RASPA2/pull/13.

— Reply to this email directly or view it on GitHub https://github.com/numat/RASPA2/pull/13#event-165627502.

patrickfuller commented 9 years ago

Hi Ernesto,

Have you been introduced yet to David Dubbeldam? He’s the author of RASPA and is probably the only one out there who understands the source. If you haven’t, I could send an intro email.

Regarding the unit cell issue, I had that same problem with FeMIL-100. The problem was due to the memory requirements for RASPA data structures being larger than the available memory. My short term fix was to use a high-memory Amazon EC2 instance, but the right way to do it would be to go through the source and make more efficient data structures.

Hope this helps, Pat