Böhme, David ; Geimer, Markus ; Wolf, Felix ; Arnold, Lukas (2016)
Identifying the Root Causes of Wait States in Large-Scale Parallel Applications.
In: ACM Transactions on Parallel Computing, 3 (2)
doi: 10.1145/2934661
Artikel, Bibliographie
Kurzbeschreibung (Abstract)
Driven by growing application requirements and accelerated by current trends in microprocessor design, the number of processor cores on modern supercomputers is increasing from generation to generation. However, load or communication imbalance prevents many codes from taking advantage of the available parallelism, as delays of single processes may spread wait states across the entire machine. Moreover, when employing complex point-to-point communication patterns, wait states may propagate along far-reaching cause-effect chains that are hard to track manually and that complicate an assessment of the actual costs of an imbalance. Building on earlier work by Meira, Jr., et al., we present a scalable approach that identifies program wait states and attributes their costs in terms of resource waste to their original cause. By replaying event traces in parallel both forward and backward, we can identify the processes and call paths responsible for the most severe imbalances, even for runs with hundreds of thousands of processes.
| Typ des Eintrags: | Artikel |
|---|---|
| Erschienen: | 2016 |
| Autor(en): | Böhme, David ; Geimer, Markus ; Wolf, Felix ; Arnold, Lukas |
| Art des Eintrags: | Bibliographie |
| Titel: | Identifying the Root Causes of Wait States in Large-Scale Parallel Applications |
| Sprache: | Englisch |
| Publikationsjahr: | 20 Juli 2016 |
| Verlag: | ACM |
| Titel der Zeitschrift, Zeitung oder Schriftenreihe: | ACM Transactions on Parallel Computing |
| Jahrgang/Volume einer Zeitschrift: | 3 |
| (Heft-)Nummer: | 2 |
| Buchtitel: | Proc. of |
| Veranstaltungstitel: | Proc. of the 39th International Conference on Parallel Processing (ICPP), San Diego, CA, USA |
| DOI: | 10.1145/2934661 |
| Kurzbeschreibung (Abstract): | Driven by growing application requirements and accelerated by current trends in microprocessor design, the number of processor cores on modern supercomputers is increasing from generation to generation. However, load or communication imbalance prevents many codes from taking advantage of the available parallelism, as delays of single processes may spread wait states across the entire machine. Moreover, when employing complex point-to-point communication patterns, wait states may propagate along far-reaching cause-effect chains that are hard to track manually and that complicate an assessment of the actual costs of an imbalance. Building on earlier work by Meira, Jr., et al., we present a scalable approach that identifies program wait states and attributes their costs in terms of resource waste to their original cause. By replaying event traces in parallel both forward and backward, we can identify the processes and call paths responsible for the most severe imbalances, even for runs with hundreds of thousands of processes. |
| Zusätzliche Informationen: | Art.No.: 11 |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Parallele Programmierung |
| Hinterlegungsdatum: | 20 Apr 2018 09:35 |
| Letzte Änderung: | 17 Mai 2024 07:13 |
| PPN: | 518391043 |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum