Training

Training Events

4 Jun 2011

Tutorial in Parallel Programming with MPI/SMPSs

25th International Conference on Supercomputing Tucson, Arizona http://ics11.cs.arizona.edu/ Abstract: MPI/SMPSs is a hybrid parallel programming model for large-scale homogeneous or heterogeneous systems. The SMPSs component of the model enables incremental parallelization of sequential code, while providing the necessary support for extracting task-level parallelism from applications at runtime, asynchronous execution, heterogeneity, modularity, and portability. Like OpenMP, SMPSs is based on directives and its tasking model can be seen as an extension of the OpenMP task...

Presentations and documentation:

Below you can download some of the slide decks used in the ICS'11 tutorial and other training material:

The files in this section contain training material.

In the following ftp site you can obtain an OVA file which includes a Linux virtual machine with installed and ready to use versions of SMPSs, Ayudame/Temanejo, Paraver and Scalasca. Examples and a simple hands-on is also included.
- FTP site: https://bts.bsc.es/uUay4LNy
- Username: uUay4LNy
  Password: 5BBAh%I3
You can check the correctness of the file with the following md5 sum: b97daa9f338066ad36dedf8b6c1a45ad
The file Welcome_to_VM.pdf attached below has instructions on how to installuse this virtual machine.

Attachment	Size
Hands on analysis of MPI/SMPSs traces This file contains some traces, cfgs and guidelines on how to analyze MPI/SMPSs traces. It uses the MPI/SMPSs version of ScaLAPACK cholesky by UJI	25.06 MB
Welcome_to_VM.pdf	237.43 KB

Training

Training Events

Presentations and documentation:

Training

Project Track

Login