Graph Repartitioning with both Dynamic Load and Dynamic Processor Allocation

Abstract : Dynamic load balancing is an important step conditioning the performance of parallel programs, like adaptive mesh refinement codes. If the global workload varies drastically over time (such that memory is exceeded), it can be relevant to adjust the number of processors while maintaining the load balanced. We propose two different solutions, that extend classic graph repartitioning approaches to accept a variable number of processors: one based on biased partitioning method and one based on a diffusive method. We call this problem: the MxN graph repartitioning problem. Finally, an experimental study on real-life graphs validates our algorithms against state-of-the-art methods.
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download

https://hal.inria.fr/hal-00857881
Contributor : Aurélien Esnard <>
Submitted on : Wednesday, September 4, 2013 - 5:03:54 PM
Last modification on : Thursday, January 11, 2018 - 6:22:35 AM

File

article-final-hal.pdf
Files produced by the author(s)

Identifiers

Citation

Clément Vuchener, Aurélien Esnard. Graph Repartitioning with both Dynamic Load and Dynamic Processor Allocation. International Conference on Parallel Computing - ParCo2013, Sep 2013, München, Germany. pp.243-252, ⟨10.3233/978-1-61499-381-0-243⟩. ⟨hal-00857881v2⟩

Share

Metrics

Record views

409

Files downloads

388