Newer
Older
[Scalasca](http://www.scalasca.org/) is a software tool that supports the performance optimization of parallel programs by measuring and analyzing their runtime behavior. The analysis identifies potential performance bottlenecks – in particular those concerning communication and synchronization – and offers guidance in exploring their causes.
Scalasca supports profiling of MPI, OpenMP and hybrid MPI+OpenMP applications.
There are currently two versions of Scalasca 2.0 [modules](../../environment-and-modules/) installed on Anselm:
- scalasca2/2.0-gcc-openmpi, for usage with [GNU Compiler](../compilers/) and [OpenMPI](../mpi/Running_OpenMPI/),
- scalasca2/2.0-icc-impi, for usage with [Intel Compiler](../compilers.html) and [Intel MPI](../mpi/running-mpich2/).
Profiling a parallel application with Scalasca consists of three steps:
1. Instrumentation, compiling the application such way, that the profiling data can be generated.
2. Runtime measurement, running the application with the Scalasca profiler to collect performance data.
Instrumentation via " scalasca -instrument" is discouraged. Use [Score-P instrumentation](score-p/).
After the application is instrumented, runtime measurement can be performed with the `scalasca -analyze` command. The syntax is:
`scalasca -analyze [scalasca options][launcher] [launcher options][program] [program options]`
- **-t Enable trace data collection. By default, only summary data are collected.**
- **-e <directory> Specify a directory to save the collected data to. By default, Scalasca saves the data to a directory with prefix scorep\_, followed by name of the executable and launch configuration.**
Scalasca can generate a huge amount of data, especially if tracing is enabled. Please consider saving the data to a [scratch directory](../../storage/storage/).
For the analysis, you must have [Score-P](score-p/) and [CUBE](cube/) modules loaded. The analysis is done in two steps, first, the data is preprocessed and then CUBE GUI tool is launched.
If you do not wish to launch the GUI tool, use the "-s" option :
Alternatively you can open CUBE and load the data directly from here. Keep in mind that in that case the preprocessing is not done and not all metrics will be shown in the viewer.
Refer to [CUBE documentation](cube/) on usage of the GUI viewer.