I have meep currently running on a computational cluster with 128 core and 785 GBytes of Ram. I'm trying to do a simulation of a WGM "sphere" coupled to a fiber. I need to run the model at a high resolution relative to the sphere diameter 1) so the fiber is accurate, 2) so the gap between the fiber and the sphere is accurate and 3) because we will be deforming the sphere into odd shapes with a spatial resolution of 50nm. My current simulation is a cell that's 501X501*501 with a resolution of 10. This should result in a mesh with ~125 Billion points. While this is a lot (a whole lot), it should be feasible on the computer cluster I am using. I'm getting the 'Cannot split -1134630259 into 128 parts' error. If I run this with a resolution of 1 it will complete without error but this won't give me the resolution I need. It seems as if there is a int overflow error occurring, but since I'm running this on a 64 bit computer, and compiled it as a 64 bit application, the integer limit should be 18,446,744,073,709,551,616, which is orders of magnitude larger than my simulation size. Is there something I'm missing here?
Thanks, Nathan
_______________________________________________ meep-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
