Chapter 19

Programming a Heterogeneous Computing Cluster

With Special Contributions fromIsaac Gelado and Javier Cabezas

Chapter Outline

19.1 Background

19.2 A Running Example

19.3 MPI Basics

19.4 MPI Point-to-Point Communication Types

19.5 Overlapping Computation and Communication

19.6 MPI Collective Communication

19.7 Summary

19.8 Exercises

References

So far we have focused on programming a heterogeneous computing system with one host and one device. In high-performance computing (HPC), many applications require the aggregate computing power of a cluster of computing nodes. Many of the HPC clusters today have one or more hosts and one or more devices in each node. Historically, these clusters have been programmed predominately with the Message ...

Get Programming Massively Parallel Processors, 2nd Edition now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.