Scheduling for parallel processing (Q1010955)

The book presents in a systematical way the interdisciplinary subject of scheduling for parallel processing. The perspective considered in presenting the concepts is that of the scheduling models. In this context the problems and algorithms encountered in scheduling processes are defined based on several basic models. Furthermore, the author goes deeper in the computational complexity theory associated with scheduling problems. The content of the book is organized in eight chapters, the first three presenting the basic notions, the next four examples of scheduling models, and the last one the applicability of scheduling models. In particular the second chapter introduces the notions of graph theory needed for scheduling models, as well as essential elements of the methodology of computational complexity theory. The fourth chapter discusses the classical deterministic scheduling theory, the fifth chapter the parallel task model, the sixth chapter scheduling with communication delays, and the seventh chapter scheduling divisible loads. The text is addressed to researchers in parallel computing and applied mathematics. It can be used for advanced lectures in parallel computing. The algorithms that are presented may be interesting for developers of parallel applications. Many of them are based on heuristics for solving hard combinatorial problems.