Commit e90739a5 authored by David Hrbáč's avatar David Hrbáč

Links OK

parent fa7cb2d2
Pipeline #5204 failed with stages
in 1 minute and 21 seconds
......@@ -2,16 +2,16 @@
## Introduction
[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][a] 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.
## Installed Versions
There are currently two versions of Scalasca 2.0 [modules](modules-matrix/) installed on Anselm:
There are currently two versions of Scalasca 2.0 [modules][1] installed on Anselm:
* scalasca2/2.0-gcc-openmpi, for usage with [GNU Compiler](software/compilers/) and [OpenMPI](software/mpi/Running_OpenMPI/),
* scalasca2/2.0-icc-impi, for usage with [Intel Compiler](software/compilers/) and [Intel MPI](software/mpi/running-mpich2/).
* scalasca2/2.0-gcc-openmpi, for usage with [GNU Compiler][2] and [OpenMPI][3],
* scalasca2/2.0-icc-impi, for usage with [Intel Compiler][2] and [Intel MPI][4].
## Usage
......@@ -23,7 +23,7 @@ Profiling a parallel application with Scalasca consists of three steps:
### Instrumentation
Instrumentation via `scalasca -instrument` is discouraged. Use [Score-P instrumentation](software/debuggers/score-p/).
Instrumentation via `scalasca -instrument` is discouraged. Use [Score-P instrumentation][5].
### Runtime Measurement
......@@ -43,11 +43,11 @@ Some notable Scalasca options are:
* **-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.**
!!! note
Scalasca can generate a huge amount of data, especially if tracing is enabled. Please consider saving the data to a [scratch directory](salomon/storage/).
Scalasca can generate a huge amount of data, especially if tracing is enabled. Please consider saving the data to a [scratch directory][6].
### Analysis of Reports
For the analysis, you must have [Score-P](software/debuggers/score-p/) and [CUBE](software/debuggers/cube/) modules loaded. The analysis is done in two steps, first, the data is preprocessed and then CUBE GUI tool is launched.
For the analysis, you must have [Score-P][5] and [CUBE][7] modules loaded. The analysis is done in two steps, first, the data is preprocessed and then CUBE GUI tool is launched.
To launch the analysis, run :
......@@ -63,8 +63,18 @@ scalasca -examine -s <experiment_directory>
Alternatively you can open CUBE and load the data directly from here. Keep in mind that in that case the pre-processing is not done and not all metrics will be shown in the viewer.
Refer to [CUBE documentation](software/debuggers/cube/) on usage of the GUI viewer.
Refer to [CUBE documentation][7] on usage of the GUI viewer.
## References
1. [http://www.scalasca.org/](http://www.scalasca.org/)
1. [http://www.scalasca.org/][a]
[1]: ../../modules-matrix.mdl
[2]: ../compilers.md
[3]: ../mpi/Running_OpenMPI.md
[4]: ../mpi/running-mpich2.md
[5]: score-p.md
[6]: ../../salomon/storage.md
[7]: cube.md
[a]: http://www.scalasca.org/
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment