StandAloneBSSN Benchmarking Application

Description

The StandAloneBSSN benchmark application is the computational kernel for the state of the art black hole simulations performed by the numerical relativity group at the Albert Einstein Institute (Max Planck Institute for Gravitational Physics). Their production simulations are typically run across hundreds of processors for days at a time.

This benchmark contains only the main computational kernel of the BenchBSSN benchmark, and is run outside of Cactus. This way one can easily run tests on it, without having to build the Cactus flesh and other Cactus thorns. It is meant to be useful to compiler developers, to illustrate the current needs of high performance computing in the numerical relativity community.

Information

• The source code for the current version, 1.0, is packaged in a gzipped tar file.
• To run the code, simply edit the Makefile for your architecture, and type 'make'. You can edit parameters in wrap.h if you like.

Floating Point Operation Count

Floating point operations are counted using perfex on our Origin 2000 (using perfex -e 21). The origin counts a multiply and add as a single operation, so we list numbers with this switched both on and off (compile with -TARG:madd=OFF to switch it off).