Presenter(s):Markus Geimer, Andreas Knüpfer, Rick Kufrin, Shirley Moore, Sameer Shende
ABSTRACT: This tutorial presents state-of-the-art performance tools for leading-edge HPC systems, demonstrating how they can be used for performance engineering of effective scientific applications. Parallel performance evaluation toolsets from the POINT (Performance Productivity through Open Integrated Tools) and VI-HPS (Virtual Institute – High Productivity Supercomputing) projects are introduced – PerfSuite, Scalasca, Vampir and TAU. We cover all aspects of performance engineering practice, including instrumentation, measurement (profiling and tracing, timing and PAPI hardware counters), data storage, analysis, and visualization. Emphasis is placed on how tools are used in combination for identifying performance problems and investigating optimization alternatives, illustrated with case studies from POINT and VI-HPS partners using key application codes. Prominently featured are hands-on exercises with participants' notebook computers using a Live-DVD containing all of the tools, which will help to prepare participants to apply modern methods for locating and diagnosing typical performance bottlenecks in real-world parallel programs at scale.
Markus Geimer - Juelich Supercomputing Centre
Andreas Knüpfer - Technical University Dresden
Rick Kufrin - National Center for Supercomputing Applications
Shirley Moore - University of Tennessee, Knoxville