Possible Memory Leak when using HSE for Molecular Dynamics
Posted: Thu Dec 13, 2012 12:46 am
I have been attempting to use the HSE functional to perform molecular dynamics. I have noticed that when using both vasp 5.2.8 and 5.3.2 that the memory used by the code is increasing monotonically with the MD steps until I eventually run out of memory and the process dies. My best guess is that the memory is being allocated during FORHF and is then not deallocated. I have tried repeating the calculation using PBE and the memory problems go away. I have also looked through the source, specifically at the routine FOCK_FORCE, but I cannot find any issues. Specifically, I have looked and all allocatable arrays in FOCK_FORCE appear to be deallocated. Does anyone have any suggestions as to how I can debug this?
Thanks
Luke Shulenburger
Thanks
Luke Shulenburger