Chapter 9. Programming a Parallel Cluster
At its most basic level, a parallel cluster is designed to make use of several processors at once and pass code between them. The combination of nodes, networks, and the monitoring software that runs in integration with each other is only one aspect of running a parallel cluster.
Remember that a parallel cluster is designed to handle a larger data set of code than a single processor system. Where one CPU can handle a small amount of code, many processors can handle that much more processing. However, in actual practice, the code has to be optimized to handle the extra processors. Although Linux can easily ...
Get Linux Clustering: Building and Maintaining Linux Clusters 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.
