Cactus Benchmarks
This page links to benchmarks for applications using Cactus,
run over a variety of machines and environments. Note that
some of these applications are constantly evolving, the date
of the benchmark can usually be found by clicking on the
number for the single processor time which links to a log
file for the actual run.
You can view recent Cactus benchmarks from the benchmark database
here.
BenchBSSN_MoL
The BenchBSSN_MoL benchmark
application is the computational kernel for the state of the
art black hole simulations performed by the numerical
relativity group at the CCT (Center for Computation and Technology) & AEI (Max
Planck Institute for Gravitational Physics). Their
production simulations are typically run across hundreds of
processors for days at a time.
BenchADM
The BenchADM benchmark application is a
computational kernel representative of many evolution
methods used in numerical relativity. It is characterised by
being compute intensive, with around 600 Floating point
operations per grid point and with six grid functions
being synchronised at each iteration. This benchmark uses a
fixed constant load on each processor.
BenchBSSN
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.
StandAloneBSSN
The StandAloneBSSN benchmark
application runs the same computational kernel as the
BenchBSSN benchmark, but runs outside of Cactus. This makes
it convenient for study, for example by compiler developers.
BenchIO
The BenchIO benchmark application
measures the speed at which large amounts of data (e.g. for
checkpointing) can be written using different IO methods. Note that
IO is highly dependent on the IO parameters chosen and the file
system and networks of the particular machine.
|
![[MAINIMAGE]](../images/development.jpg)
