Efficient Multiple Processor Coordination.

Abstract

In recent years the concept of level-structured system organization has received growing acceptance as a computer operating system design technique. This approach treats the hardware of a computing system as the bottom level of a multilevel system which will be built up one level at a time. Each level creats an abstraction of the next lower level by implementing a new virtual machine which provides some features not previously available. This dissertation presents efficient solutions to certain problems of processor interaction which must be faced by the system designer at the lower levels of a strictly level-structured operating system. A loop-free algorithm is developed which solves Dijkstra's critical section problem, based on a rudimentary processor delay mechanism and the instruction set of a typical large computer of conventional architecture (DECsystem-10). A completely rigorous proof of the algorithm's correctness is given. The algorithm is used as the basis of a systematic procedure for obtaining efficient solutions to general processor coordination problems. The procedure is illustrated by solving several of the well-known reader/writer problems and a more complex new problem, the producer/consumer/distributor problem, which involves simultaneous use of several independent shared resources. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 1976
Accession Number
ADA032803

Entities

People

  • Bob E. Baker

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • C4I
  • Energy and Power Technologies
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Air Force
  • Assembly Languages
  • Central Processing Units
  • Computer Architecture
  • Computer Programming
  • Computer Programs
  • Computers
  • Computing System Architectures
  • Data Storage Systems
  • Databases
  • Engineering
  • Instruction Set Architecture
  • Language
  • Multiprogramming
  • Operating Systems
  • Programming Languages
  • Virtual Machines

Readers

  • Operations Research
  • Parallel and Distributed Computing.
  • Systems Analysis and Design