Publication | Closed Access
The Potential of Diffusive Load Balancing at Large Scale
11
Citations
12
References
2016
Year
Unknown Venue
Cluster ComputingLoad Balancing (Computing)EngineeringComputer ArchitectureNetwork AnalysisCloud Load BalancingParallel MetaheuristicsNeighbor NodesSystems EngineeringParallel ComputingCombinatorial OptimizationMassively-parallel ComputingLoad BalancingLarge ScaleComputer EngineeringDiffusive MethodsComputer ScienceDynamic LoadDistributed ProcessingGraph TheoryLoad ShiftingEdge ComputingCloud ComputingParallel Programming
Dynamic load balancing with diffusive methods is known to provide minimal load transfer and requires communication between neighbor nodes only. These are very attractive properties for highly parallel systems. We compare diffusive methods with state-of-the-art geometrical and graph-based partitioning methods on thousands of nodes. When load balancing overheads, i.e. repartitioning computation time and migration, have to be minimized, diffusive methods provide substantial benefits.
| Year | Citations | |
|---|---|---|
Page 1
Page 1