Potentials and limitations of fault-based Markov prefetching for virtual memory pages


Journal Article

Fault-based Markov prefetching for virtual memory pages (VMP) is examined. Markov prediction based only on the sequence of program-issued faults are shown to achieve reasonably high levels of accuracy for some scientific applications. Using the fault sequence to predict future faults makes speculative prefetching of VMPs feasible to implement. Further, despite high prediction accuracy, it is difficult in practice to obtain speedup from one-fault ahead speculative prefetching as this approach requires fault rate and fault latency sufficiently low that I/O overhead is not prohibitive in the first place.

Full Text

Duke Authors

Cited Authors

  • Bartels, G; Karlin, A; Levy, H; Voelker, G; Anderson, D; Chase, J

Published Date

  • January 1, 1999

Published In

Volume / Issue

  • 27 / 1

Start / End Page

  • 206 - 207

International Standard Serial Number (ISSN)

  • 0163-5999

Digital Object Identifier (DOI)

  • 10.1145/301464.301572

Citation Source

  • Scopus