An Analytical Cache Model.

Abstract

Trace driven simulation and hardware measurement are the techniques most often used to obtain accurate performance figures for caches. The former requires a large amount of simulation time to evaluate each cache configuration while the latter is restricted to measurements of existing caches. An analytical cache model that uses parameters extracted from address traces of programs can provide estimates of cache performance and show the effects of varying cache parameters. By representing the factors that affect cache performance, the authors develop an analytical model that gives miss rates for a given trace as a function of cache-size, degree of associativity, block-size multiprogramming level, task switch interval, and observation interval. The predicted values closely approximate the results of trace drive simulations while requiring only a small fraction of the computation cost. Keywords: run length; start up.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1986
Accession Number
ADA181716

Entities

People

  • Anant Agarwal
  • John Hennessy
  • Mark Horowitz

Organizations

  • Stanford University

Tags

DTIC Thesaurus Topics

  • Accuracy
  • Boundaries
  • Collisions
  • Computations
  • Computer Programming
  • Computers
  • Equations
  • Errors
  • Intervals
  • Markov Models
  • Measurement
  • Models
  • Multiprogramming
  • Probability
  • Simulations
  • Statistics
  • Steady State

Readers

  • Parallel and Distributed Computing.
  • Systems Analysis and Design