A Taxonomy for Describing Matching and Scheduling Heuristics for Mixed-Machine Heterogeneous Computing Systems

The problem of mapping (deened as matching and scheduling) tasks and communications onto multiple machines and networks in a heterogeneous computing (HC) environment has been shown to be NP-complete, in general, requiring the development of heuristic techniques. Many diierent types of mapping heuristics have been developed in recent years. However, selecting the best heuristic to use in any given scenario remains a diicult problem. Factors making this selection diicult are discussed. Motivated by these diicul-ties, a new taxonomy for classifying mapping heuris-tics for HC environments is proposed (\the Purdue HC Taxonomy"). The taxonomy is deened in three major parts: (1) the models used for applications and communication requests, (2) the models used for target hardware platforms, and (3) the characteristics of mapping heuristics. Each part of the taxonomy is described, with examples given to help clarify the taxonomy. The ben-eets and uses of this taxonomy are also discussed.