Analytic Models in Memory and Processor Allocation.

Abstract

A problem of memory and processor allocation to competing programs in time-sharing environment is investigated. Two level (core and auxiliary) paged memory system is studied. An analytic model of CPU time needed (instructions executed) and memory required (pages needed) is proposed as an approximation to actual program behavior. Specific probability distributions are postulated for CPU time needed and memory required; these are used in studying memory and processor allocation. Program behavior (execution) in paging environment is modeled; a memory-use estimator function is derived (page size dependence of this function is explicitly included). This model conforms to previous results established in simulation studies of program behavior. Based on memory-use estimator function, paging characteristics of a single program and the system under multiprogramming are studied. (Author Modified Abstract)

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 1972
Accession Number
AD0756493

Entities

People

  • Jai Ram Rao

Organizations

  • University of California, Los Angeles

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Environment
  • Estimators
  • Instructions
  • Mathematics
  • Multiprogramming
  • Probability
  • Probability Distributions
  • Random Variables
  • Simulations

Fields of Study

  • Computer science

Readers

  • Computational Modeling and Simulation
  • Parallel and Distributed Computing.
  • Statistical inference.