Measuring application performance often results in a large amount of profile data or traces that are difficult to handle or interpret beyond some trivial first observations. These analyses often do not provide the kind of insight that would really help a code developer determine the most appropriate direction to follow in order to improve the code
Is it possible to compute a very limited number of metrics for a parallel MPI application that explains its behaviour in terms of fundamental properties?
This webinar introduced and explained the model being used in the POP Project that achieves such high semantic description of an application efficiency with just three metrics at the first level. It also demonstrated how these metrics can be easily computed with the BSC tools and Scalasca toolset as well as how the top three metrics can be derived from a standard MPI profiler output. It concluded by showing metrics for some example applications and how different behavioural effects of a system or application are reflected in the metrics.
About the Presenter
Jesus Labarta is full professor of Computer Architecture at the Technical University of Catalonia (UPC) since 1990. Since 2005, he has been responsible for the Computer Science Research Department within the Barcelona Supercomputing Center (BSC). His major directions of current work relate to performance analysis tools, programming models and resource management. His team distributes the Open Source BSC tools (Paraver and Dimemas) and performs research on increasing the intelligence embedded in the performance analysis tools. He is involved in the development of the OmpSs programming model and its different implementations for SMP, GPUs and cluster platforms.