Skip to main navigation Skip to search Skip to main content

An Experimental Study to Determine Task Size for Rollback Recovery Systems

  • University of Wisconsin-Madison

Research output: Contribution to journalArticlepeer-review

11 Scopus citations

Abstract

The effects of using a recovery cache to save the variables of a program are studied. A new optimization model for rollback is formulated to include the effects of a recovery cache in rollback systems. The parameters of the model proposed in this correspondence are the maximum recovery time, the cache size, and the save and load times associated with the task size. We also discuss the results of an experimental study conducted to estimate the parameters of the programs that are critical for arriving at a suitable task size or cache size to minimize the cost of recovery.

Original languageEnglish
Pages (from-to)872-877
Number of pages6
JournalIEEE Transactions on Computers
Volume37
Issue number7
DOIs
StatePublished - Jul 1988

Keywords

  • Program graph
  • recovery
  • recovery cache
  • recovery time
  • rollback
  • task size

Fingerprint

Dive into the research topics of 'An Experimental Study to Determine Task Size for Rollback Recovery Systems'. Together they form a unique fingerprint.

Cite this