O'Reilly logo

Automatic Parallelization by Samuel P. Midkiff

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

CHAPTER 6

Compiling for distributed memory machines

Up until now, our focus has been almost entirely on compiling for parallelism on shared memory machines, motivated in part by the dominance of multicore processors. Nevertheless, all large machines for high-performance numerical computing, as well as many small scale clusters, have a physically distributed memory architecture and are programmed using a distributed memory programming model.

Distributed memory machines consists of nodes connected to one another by using Ethernet or a variety of proprietary interfaces. Examples of these include the IBM Blue Gene machines [29, 162], clusters of workstations and small clusters of rack mounted nodes [173]. Distributed memory machines usually execute ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required