Highly Concurrent Scalar Processing.

Abstract

High speed scalar processing is an essential characteristic of high performance general purpose computer systems. Efficient concurrent execution of scalar code is difficult due to data dependencies and conditional branches. This thesis proposes a code scheduling heuristic call the decision tree scheduling (DTS) technique for general scalar code, and an optimal code scheduling algorithm called the simple loop scheduling (SLS) algorithm for a restricted class of innermost loops. Also proposed is a highly concurrent machine architecture that takes advantage of these scheduling techniques. Keywords: parallel processing; throughput; instructions; multiprocessors.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1986
Accession Number
ADA163195

Entities

People

  • Peter Y. Hsu

Organizations

  • University of Illinois Urbana–Champaign

Tags

Communities of Interest

  • C4I
  • Energy and Power Technologies
  • Human Systems

DTIC Thesaurus Topics

  • Algorithms
  • Application Software
  • Computer Architecture
  • Computer Programming
  • Computer Science
  • Computers
  • Floating Point Operations
  • Instruction Set Architecture
  • Operating Systems
  • Parallel Processing
  • Programming Languages
  • Scheduling (Production)
  • Semantic Models
  • Two Dimensional
  • Universities
  • Web Browsers
  • Workload

Fields of Study

  • Computer science

Readers

  • Parallel and Distributed Computing.