An Overview of Process Mapping Techniques and Algorithms in High-Performance Computing

Torsten Hoefler (1) , Emmanuel Jeannot (2, 3) , Guillaume Mercier (4, 2, 3)

1 D-INFK - Department of Computer Science [ETH Zürich]
2 RUNTIME - Efficient runtime systems for parallel architectures
3 LaBRI - Laboratoire Bordelais de Recherche en Informatique
4 Institut Polytechnique de Bordeaux

Due to the advent of modern hardware architectures of high-performance comput- ers, the way the parallel applications are laid out is of paramount importance for performance. This chapter surveys several techniques and algorithms that efficiently address this issue: the mapping of the application's virtual topology (for instance its communication pattern) onto the physical topology. Using such strategy enables to improve the application overall execution time significantly. The chapter concludes by listing a series of open issues and problems.

Domains: Distributed, Parallel, and Cluster Computing [cs.DC]

Torsten Hoefler, Emmanuel Jeannot, Guillaume Mercier. An Overview of Process Mapping Techniques and Algorithms in High-Performance Computing. Emmanuel Jeannot and Julius Zilinskas. High Performance Computing on Complex Environments, Wiley, pp.75-94, 2014, 978-1-118-71205-4. ⟨hal-00921626⟩