Optimizing parallel reduction operations

PDF Version Also Available for Download.

Description

A parallel program consists of sets of concurrent and sequential tasks. Often, a reduction (such as array sum) sequentially combines values produced by a parallel computation. Because reductions occur so frequently in otherwise parallel programs, they are good candidates for optimization. Since reductions may introduce dependencies, most languages separate computation and reduction. The Sisal functional language is unique in that reduction is a natural consequence of loop expressions; the parallelism is implicit in the language. Unfortunately, the original language supports only seven reduction operations. To generalize these expressions, the Sisal 90 definition adds user-defined reductions at the language level. Applicable ... continued below

Physical Description

50 p.

Creation Information

Denton, S.M. June 1, 1995.

Context

This thesis or dissertation is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided by UNT Libraries Government Documents Department to Digital Library, a digital repository hosted by the UNT Libraries. It has been viewed 13 times . More information about this document can be viewed below.

Who

People and organizations associated with either the creation of this thesis or dissertation or its content.

Author

Sponsor

Publisher

Provided By

UNT Libraries Government Documents Department

Serving as both a federal and a state depository library, the UNT Libraries Government Documents Department maintains millions of items in a variety of formats. The department is a member of the FDLP Content Partnerships Program and an Affiliated Archive of the National Archives.

Contact Us

What

Descriptive information to help identify this thesis or dissertation. Follow the links below to find similar items on the Digital Library.

Description

A parallel program consists of sets of concurrent and sequential tasks. Often, a reduction (such as array sum) sequentially combines values produced by a parallel computation. Because reductions occur so frequently in otherwise parallel programs, they are good candidates for optimization. Since reductions may introduce dependencies, most languages separate computation and reduction. The Sisal functional language is unique in that reduction is a natural consequence of loop expressions; the parallelism is implicit in the language. Unfortunately, the original language supports only seven reduction operations. To generalize these expressions, the Sisal 90 definition adds user-defined reductions at the language level. Applicable optimizations depend upon the mathematical properties of the reduction. Compilation and execution speed, synchronization overhead, memory use and maximum size influence the final implementation. This paper (1) Defines reduction syntax and compares with traditional concurrent methods; (2) Defines classes of reduction operations; (3) Develops analysis of classes for optimized concurrency; (4) Incorporates reductions into Sisal 1.2 and Sisal 90; (5) Evaluates performance and size of the implementations.

Physical Description

50 p.

Notes

OSTI as DE96003838

Source

  • Other Information: TH: Thesis (M.S.)

Language

Identifier

Unique identifying numbers for this document in the Digital Library or other systems.

  • Other: DE96003838
  • Report No.: UCRL-LR--122354
  • Grant Number: W-7405-ENG-48
  • Office of Scientific & Technical Information Report Number: 188648
  • Archival Resource Key: ark:/67531/metadc671303

Collections

This document is part of the following collection of related materials.

Office of Scientific & Technical Information Technical Reports

Reports, articles and other documents harvested from the Office of Scientific and Technical Information.

Office of Scientific and Technical Information (OSTI) is the Department of Energy (DOE) office that collects, preserves, and disseminates DOE-sponsored research and development (R&D) results that are the outcomes of R&D projects or other funded activities at DOE labs and facilities nationwide and grantees at universities and other institutions.

What responsibilities do I have when using this thesis or dissertation?

When

Dates and time periods associated with this thesis or dissertation.

Creation Date

  • June 1, 1995

Added to The UNT Digital Library

  • June 29, 2015, 9:42 p.m.

Description Last Updated

  • Feb. 19, 2016, 8:31 p.m.

Usage Statistics

When was this document last used?

Yesterday: 0
Past 30 days: 0
Total Uses: 13

Interact With This Thesis Or Dissertation

Here are some suggestions for what to do next.

Start Reading

PDF Version Also Available for Download.

International Image Interoperability Framework

IIF Logo

We support the IIIF Presentation API

Denton, S.M. Optimizing parallel reduction operations, thesis or dissertation, June 1, 1995; California. (digital.library.unt.edu/ark:/67531/metadc671303/: accessed September 22, 2018), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT Libraries Government Documents Department.