adithyarao3103 / SYM1DMMMT

Simulation code for the matrix model of M-theory. It includes the BFSS model and the BMN plane wave deformation. Two combined versions of the code. One code by Masanori Hanada and second a GPU version with some additional optimizations. Programming language: Fortran 90/95 + OpenACC + CUDA. The most recent version includes also the MPI code by Masan

SYM1DMMMT

Current version is a development version of the code.

Compile

The configuration is done with a small script ./configure.sh. This allows to select between different version of the code

1. serial version (one core CPU), forked from Masanori Hanada and updated

2. GPU version (CUDA + OpenAcc)

3. matrix parallel (MPI), forked from Masanori Hanada and updated

There are still a view settings that have to be manually edited in the Makefile.inc

Compile time parameters

In the current version of the code several parameters have to be set manually. These are in the file

staticparameters.f90

The MPI code has a separate file

matrix_parallel/size_parallel.h

Testruns

These can be found in the subfolders example*.