BenchBSSN Benchmarking Application
Description
The BenchBSSN 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 uses some thorns which are not part of the
Cactus distribution, but are freely available from the AEI
CVS repository. Note that this evolution scheme is still
being enhanced and developed and so benchmark numbers may
change somewhat in time.
Information
- ThornList for BenchBSSN: BenchBSSN.th
- Parameter file for BenchBSSN: In
AEIThorns/ADM_BSSN thorn,
ADM_BSSN/par/BenchBSSN_40l.par and ADM_BSSN/par/BenchBSSN_80l.par
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).
|
|
40x40x20
|
80x80x40
|
|
Number of grid points
|
32,000
|
256,000
|
|
Standard optimisations and madd=OFF
|
28,620,205,421
|
243,454,608,991
|
|
Standard optimisations and madd=ON
|
|
|
No optimisations and madd=OFF
|
|
|
|