On the Use of Probabilistic Worst-Case Execution Time Estimation for Parallel Applications in High Performance Systems
Some high performance computing (HPC) applications exhibit increasing real-time requirements, which call for effective means to predict their high execution times distribution. This is a new challenge for HPC applications but a well-known problem for real-time embedded applications where solutions a...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-03-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/8/3/314 |
| _version_ | 1818298374898581504 |
|---|---|
| author | Matteo Fusi Fabio Mazzocchetti Albert Farres Leonidas Kosmidis Ramon Canal Francisco J. Cazorla Jaume Abella |
| author_facet | Matteo Fusi Fabio Mazzocchetti Albert Farres Leonidas Kosmidis Ramon Canal Francisco J. Cazorla Jaume Abella |
| author_sort | Matteo Fusi |
| collection | DOAJ |
| description | Some high performance computing (HPC) applications exhibit increasing real-time requirements, which call for effective means to predict their high execution times distribution. This is a new challenge for HPC applications but a well-known problem for real-time embedded applications where solutions already exist, although they target low-performance systems running single-threaded applications. In this paper, we show how some performance validation and measurement-based practices for real-time execution time prediction can be leveraged in the context of HPC applications on high-performance platforms, thus enabling reliable means to obtain real-time guarantees for those applications. In particular, the proposed methodology uses coordinately techniques that randomly explore potential timing behavior of the application together with Extreme Value Theory (EVT) to predict rare (and high) execution times to, eventually, derive probabilistic Worst-Case Execution Time (pWCET) curves. We demonstrate the effectiveness of this approach for an acoustic wave inversion application used for geophysical exploration. |
| first_indexed | 2024-12-13T04:34:19Z |
| format | Article |
| id | doaj.art-937ed55eeb124d889db9631f601ca49c |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-12-13T04:34:19Z |
| publishDate | 2020-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-937ed55eeb124d889db9631f601ca49c2022-12-21T23:59:28ZengMDPI AGMathematics2227-73902020-03-018331410.3390/math8030314math8030314On the Use of Probabilistic Worst-Case Execution Time Estimation for Parallel Applications in High Performance SystemsMatteo Fusi0Fabio Mazzocchetti1Albert Farres2Leonidas Kosmidis3Ramon Canal4Francisco J. Cazorla5Jaume Abella6Barcelona Supercomputing Center (BSC), Cr. Jordi Girona 31, 08034 Barcelona, SpainBarcelona Supercomputing Center (BSC), Cr. Jordi Girona 31, 08034 Barcelona, SpainBarcelona Supercomputing Center (BSC), Cr. Jordi Girona 31, 08034 Barcelona, SpainBarcelona Supercomputing Center (BSC), Cr. Jordi Girona 31, 08034 Barcelona, SpainBarcelona Supercomputing Center (BSC), Cr. Jordi Girona 31, 08034 Barcelona, SpainBarcelona Supercomputing Center (BSC), Cr. Jordi Girona 31, 08034 Barcelona, SpainBarcelona Supercomputing Center (BSC), Cr. Jordi Girona 31, 08034 Barcelona, SpainSome high performance computing (HPC) applications exhibit increasing real-time requirements, which call for effective means to predict their high execution times distribution. This is a new challenge for HPC applications but a well-known problem for real-time embedded applications where solutions already exist, although they target low-performance systems running single-threaded applications. In this paper, we show how some performance validation and measurement-based practices for real-time execution time prediction can be leveraged in the context of HPC applications on high-performance platforms, thus enabling reliable means to obtain real-time guarantees for those applications. In particular, the proposed methodology uses coordinately techniques that randomly explore potential timing behavior of the application together with Extreme Value Theory (EVT) to predict rare (and high) execution times to, eventually, derive probabilistic Worst-Case Execution Time (pWCET) curves. We demonstrate the effectiveness of this approach for an acoustic wave inversion application used for geophysical exploration.https://www.mdpi.com/2227-7390/8/3/314wcetprobabilistic timing analysisrandomizationmeasurement-basedhpc applications |
| spellingShingle | Matteo Fusi Fabio Mazzocchetti Albert Farres Leonidas Kosmidis Ramon Canal Francisco J. Cazorla Jaume Abella On the Use of Probabilistic Worst-Case Execution Time Estimation for Parallel Applications in High Performance Systems Mathematics wcet probabilistic timing analysis randomization measurement-based hpc applications |
| title | On the Use of Probabilistic Worst-Case Execution Time Estimation for Parallel Applications in High Performance Systems |
| title_full | On the Use of Probabilistic Worst-Case Execution Time Estimation for Parallel Applications in High Performance Systems |
| title_fullStr | On the Use of Probabilistic Worst-Case Execution Time Estimation for Parallel Applications in High Performance Systems |
| title_full_unstemmed | On the Use of Probabilistic Worst-Case Execution Time Estimation for Parallel Applications in High Performance Systems |
| title_short | On the Use of Probabilistic Worst-Case Execution Time Estimation for Parallel Applications in High Performance Systems |
| title_sort | on the use of probabilistic worst case execution time estimation for parallel applications in high performance systems |
| topic | wcet probabilistic timing analysis randomization measurement-based hpc applications |
| url | https://www.mdpi.com/2227-7390/8/3/314 |
| work_keys_str_mv | AT matteofusi ontheuseofprobabilisticworstcaseexecutiontimeestimationforparallelapplicationsinhighperformancesystems AT fabiomazzocchetti ontheuseofprobabilisticworstcaseexecutiontimeestimationforparallelapplicationsinhighperformancesystems AT albertfarres ontheuseofprobabilisticworstcaseexecutiontimeestimationforparallelapplicationsinhighperformancesystems AT leonidaskosmidis ontheuseofprobabilisticworstcaseexecutiontimeestimationforparallelapplicationsinhighperformancesystems AT ramoncanal ontheuseofprobabilisticworstcaseexecutiontimeestimationforparallelapplicationsinhighperformancesystems AT franciscojcazorla ontheuseofprobabilisticworstcaseexecutiontimeestimationforparallelapplicationsinhighperformancesystems AT jaumeabella ontheuseofprobabilisticworstcaseexecutiontimeestimationforparallelapplicationsinhighperformancesystems |