This title appears in the Scientific Report :
2021
Please use the identifier:
http://hdl.handle.net/2128/29447 in citations.
Hands-on Practical Hybrid Parallel Application Performance Engineering
Hands-on Practical Hybrid Parallel Application Performance Engineering
This tutorial presents state-of-the-art performance tools for leading-edge HPC systems founded on the community-developed Score-P instrumentation and measurement infrastructure, demonstrating how they can be used for performance engineering of effective scientific applications based on standard MPI,...
Saved in:
Personal Name(s): | Shende, Sameer |
---|---|
Geimer, Markus (Corresponding author) / Schlütter, Marc / Wesarg, Bert / Williams, Bill / Wylie, Brian J. N. (Corresponding author) | |
Contributing Institute: |
Jülich Supercomputing Center; JSC |
Imprint: |
2021
|
Conference: | The International Conference for High Performance Computing, Networking, Storage, and Analysis '21, St. Louis, MO, & Online (USA), 2021-11-14 - |
Document Type: |
Lecture |
Research Program: |
Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups |
Link: |
OpenAccess |
Publikationsportal JuSER |
This tutorial presents state-of-the-art performance tools for leading-edge HPC systems founded on the community-developed Score-P instrumentation and measurement infrastructure, demonstrating how they can be used for performance engineering of effective scientific applications based on standard MPI, OpenMP, hybrid combination of both, and increasingly common usage of accelerators. Parallel performance tools from VI-HPS.org are introduced and featured in hands-on exercises with Score-P, 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, tuning and visualization. Emphasis is placed on how tools are used together for identifying performance problems and investigating optimization alternatives. Using an AWS instance of E4S with all of the necessary tools, participants will conduct exercises with support for a remote desktop session for GUI tools. This will help to prepare participants to locate and diagnose performance bottlenecks in their own parallel programs. |