Task Scheduling for Parallel Systems (Wiley Series on Parallel and Distributed Computing)

Task Scheduling for Parallel Systems (Wiley Series on Parallel and Distributed Computing)

Publisher: Wiley-Interscience
Number Of Pages: 296
Publication Date: 2007-05-04
Sales Rank: 1573523
ISBN / ASIN: 0471735760
EAN: 9780471735762
Binding: Hardcover
Manufacturer: Wiley-Interscience
Studio: Wiley-Interscience
Average Rating: 4

A new model for task scheduling that dramatically improves the efficiency of parallel systems

Task scheduling for parallel systems can become a quagmire of heuristics, models, and methods that have been developed over the past decades. The author of this innovative text cuts through the confusion and complexity by presenting a consistent and comprehensive theoretical framework along with realistic parallel system models. These new models, based on an investigation of the concepts and principles underlying task scheduling, take into account heterogeneity, contention for communication resources, and the involvement of the processor in communications.

For readers who may be new to task scheduling, the first chapters are essential. They serve as an excellent introduction to programming parallel systems, and they place task scheduling within the context of the program parallelization process. The author then reviews the basics of graph theory, discussing the major graph models used to represent parallel programs. Next, the author introduces his task scheduling framework. He carefully explains the theoretical background of this framework and provides several examples to enable readers to fully understand how it greatly simplifies and, at the same time, enhances the ability to schedule.

The second half of the text examines both basic and advanced scheduling techniques, offering readers a thorough understanding of the principles underlying scheduling algorithms. The final two chapters address communication contention in scheduling and processor involvement in communications.

Each chapter features exercises that help readers put their new skills into practice. An extensive bibliography leads to additional information for further research. Finally, the use of figures and examples helps readers better visualize and understand complex concepts and processes.

Researchers and students in distributed and parallel computer systems will find that this text dramatically improves their ability to schedule tasks accurately and efficiently.

Review:

decompose a program into a directed graph

As computer hardware continues to improve, along the lines of Moore's Law, software improvements have lagged. This has been particularly so with a system of parallel machines. The biggest problem is how to translate a given algorithm's implementation, that was written for a single CPU, into one for several CPUs. More to the point, how to do this systematically, instead of in a manual, ad hoc fashion?

That is the promise of Sinnen's book. It takes a program, and finds a task graph. That is, the code is decomposed into a set of smaller modules. Each module can be taken to be a node of a directed graph. The directed nature of the graph is essential to the discussion. The direction of a vertex indicates a dependency on another node. Given this graph, and a set of CPUs, the book explains how to allocate the nodes across the CPUs, in a manner that is optimal.

Important related concepts are developed. Like clustering heuristics. These are used to form groups or clusters of modules. To a programmer, think of the concept of recursion. That's the utility of clustering.

The text is heavily mathematical, as befits any application of graph theory. The crucial part of the text is at its start, as you might surmise. It explains how to algorithmically extract the dependence graph, which is a representation of the dependence structure of the initial program. Potentially very useful.Password: gigapedia.org

http://rapidshare.com/files/53081008/Task_Scheduling_for_Parallel_Systems_-_0471735760.pdf.rar