UMPa: A multi-objective, multi-level partitioner for communication minimization We propose a directed hypergraph model and a refinement heuristic to distribute communicating tasks among the processing units in a distributed memory setting. The aim is to achieve load balance and minimize the maximum data sent by a processing unit. We also take two other communication metrics into account with a tie-breaking scheme. With this approach, task distributions causing an excessive use of network or a bottleneck processor which participates to almost all of the communication are avoided. par We show on a large number of problem instances that our model improves the maximum data sent by a processor up to 34% for parallel environments with 4,16,64 and 256 processing units compared to the state of the art which only minimizes the total communication volume.
Keywords for this software
References in zbMATH (referenced in 4 articles , 1 standard article )
Showing results 1 to 4 of 4.
- Smith, Cameron W.; Rasquin, Michel; Ibanez, Dan; Jansen, Kenneth E.; Shephard, Mark S.: Improving unstructured mesh partitions for multiple criteria using mesh adjacencies (2018)
- Slota, George M.; Madduri, Kamesh; Rajamanickam, Sivasankaran: Complex network partitioning using label propagation (2016)
- Sebastian Schlag, Vitali Henne, Tobias Heuer, Henning Meyerhenke, Peter Sanders, Christian Schulz: k-way Hypergraph Partitioning via n-Level Recursive Bisection (2015) arXiv
- Çatalyürek, Ümit V.; Deveci, Mehmet; Kaya, Kamer; Uçar, Bora: UMPa: A multi-objective, multi-level partitioner for communication minimization (2013)