Fast generation of large task network mappings

K.-E. Berger 1 F. Galea 2 B. Le Cun 1 R. Sirdey 2
1 PRISM - Parallélisme, Réseaux, Systèmes, Modélisation
2 DACLE-LIST - Département d'Architectures, Conception et Logiciels Embarqués-LIST
Abstract : In the context of networks of massively parallel execution models, optimizing the locality if inter-process communication is a major performance issue. We propose two heuristics to solve a dataflow process network mapping problem, where a network of communicating tasks is placed into a set of processors with limited resource capacities, while minimizing the overall communication bandwidth between processors. Those approaches are designed to tackle instances of over one hundred thousand tasks in acceptable time.
Keywords : Interprocess communication Massively parallels Process network mappings Distributed computer systems Heuristics Execution optimization Dataflow process networks Communication bandwidth Communicating tasks Parallel processing systems Optimization Mapping
Document type :
Conference papers
Complete list of metadatas
https://hal-cea.archives-ouvertes.fr/cea-01839857
Contributor : Léna Le Roy <>
Submitted on : Monday, July 16, 2018 - 10:06:01 AM
Last modification on : Wednesday, January 23, 2019 - 2:39:33 PM

Identifiers

Collections

CEA | CNRS | PRISM | UVSQ | DRT | LIST

Citation

K.-E. Berger, F. Galea, B. Le Cun, R. Sirdey. Fast generation of large task network mappings. 2014 IEEE International Parallel & Distributed Processing Symposium Workshops, May 2014, Phoenix, United States. pp.1526-1530, ⟨10.1109/IPDPSW.2014.170⟩. ⟨cea-01839857⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

54

Contact

Informations

CCSD