The study of general purpose parallel computing requires efficient and inexpensive platforms for parallel program execution. This helps in ascertaining tradeoff choices between hardware complexity and software solutions for massively parallel systems design. In this report, we present an implementation of an efficient parallel execution model on shared memory multiprocessors based on a Threaded Abstract Machine. We discuss a k-way generalized locking strategy suitable for our model. We study the performance gains obtained by a queuing strategy which uses multiple queues with reduced access contention. We also present performance models in shared memory machines, related to lock contention and serialization in shared memory allocation. A bin-based memory management technique which reduces the serialization is presented. These issues are critical for obtaining an efficient parallel execution environment.
Title
Parallel Language Support on Shared Memory Multiprocessors
Published
1991-05-22
Full Collection Name
Electrical Engineering & Computer Sciences Technical Reports
Other Identifiers
CSD-91-631
Type
Text
Extent
48 p
Archive
The Engineering Library
Usage Statement
Researchers may make free and open use of the UC Berkeley Library’s digitized public domain materials. However, some materials in our online collections may be protected by U.S. copyright law (Title 17, U.S.C.). Use or reproduction of materials protected by copyright beyond that allowed by fair use (Title 17, U.S.C. § 107) requires permission from the copyright owners. The use or reproduction of some materials may also be restricted by terms of University of California gift or purchase agreements, privacy and publicity rights, or trademark law. Responsibility for determining rights status and permissibility of any use or reproduction rests exclusively with the researcher. To learn more or make inquiries, please see our permissions policies (https://www.lib.berkeley.edu/about/permissions-policies).
