This title appears in the Scientific Report :
2015
Runtime Analysis of Parallel Applications for Industrial Software Development
Runtime Analysis of Parallel Applications for Industrial Software Development
Utilizing the parallelism offered by multicore CPUs is hard, though profiling and tracing are established techniques to understand, debug, engineer, and optimize codes. While many tools are available to capture profiles and traces, these tools are often difficult to use in industrial contexts. Tool...
Saved in:
Personal Name(s): | Feld, Christian (Corresponding author) |
---|---|
Becker, Daniel | |
Contributing Institute: |
Jülich Supercomputing Center; JSC |
Imprint: |
2015
|
Conference: | Multicore Developers Conference, Santa Clara, CA (USA), 2015-05-06 - 2015-05-07 |
Document Type: |
Poster |
Research Program: |
Computational Science and Mathematical Methods |
Publikationsportal JuSER |
Utilizing the parallelism offered by multicore CPUs is hard, though profiling and tracing are established techniques to understand, debug, engineer, and optimize codes. While many tools are available to capture profiles and traces, these tools are often difficult to use in industrial contexts. Tool development often started with sequential codes to transition to parallelism not until later, resulting in improper feature sets and usability. In contrast, parallel tools are often targeted towards HPC. As this renders these tools less suitable for codes using alternative threading models (POSIX, Qt, and ACE), this talk presents extensions to the open-source tools Score-P and Scalasca. Score-P captures detailed execution data allowing Scalasca to perform an automatic performance analysis. |