Conference Presentation (After Call)

FZJ-2014-01299

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png Hands-on Practical Hybrid Parallel Application Performance Engineering

Geimer, M. (Corresponding author)FZJ* ; Shende, S. ; Wesarg, B. ; Wylie, B. J. N.FZJ*

2013

Please use a persistent id in citations: http://hdl.handle.net/2128/5921

Abstract: This tutorial presents state-of-the-art performance tools for leading-edge HPC systems founded on the Score-P community instrumentation and measurement infrastructure, demonstrating how they can be used for performance engineering of effective scientific applications based on standard MPI, OpenMP, hybrid MPI+OpenMP, and increasingly common usage of accelerators. Parallel performance evaluation tools from the VI-HPS (Virtual Institute High Productivity Supercomputing) are introduced and featured in hands-on exercises with Scalasca, Vampir and TAU. We present the complete workflow of performance engineering, 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. Using their own notebook computers with a provided Linux Live-ISO image containing the tools (booted from DVD/USB or within a virtual machine) will help to prepare participants to locate and diagnose performance bottlenecks in their own parallel programs.

---

Contributing Institute(s):

1. Jülich Supercomputing Center (JSC)

Research Program(s):

1. 411 - Computational Science and Mathematical Methods (POF2-411) (POF2-411)
2. ATMLPP - ATML Parallel Performance (ATMLPP) (ATMLPP)

Appears in the scientific report 2013

---

Database coverage:

---