An analysis of main-memory and log space usage in extended ephemeral logging

Richard D. Regan, Alex Delis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Extended Ephemeral Logging (XEL) is a database logging and recovery technique which manages a log of recovery data by partitioning it into a series of logically circular generations. XEL copies longer-lived log data from one generation to another in order to reclaim more quickly the space occupied by shorter-lived log data. As a result of copying, records in the log lose their original ordering; this leads to main-memory and log space overhead for obsolete recovery data. In this paper, we quantify the e_ects of reordering log records by introducing the notion of Garbage Removal Dependencies (GRDs). We develop a classi_cation of log records based on GRDs and use it to characterize main-memory and log space allocation during normal system operation. Through simulation, we demonstrate how main-memory and log space allocation vary with changes in database and workload parameters.

Original languageEnglish (US)
Title of host publicationAdvances in Databases - 18th British National Conference on Databases, BNCOD 2001, Proceedings
EditorsBrian Read
PublisherSpringer Verlag
Pages24-42
Number of pages19
ISBN (Print)9783540422655
DOIs
StatePublished - 2001
Event18th British National Conference on Databases, BNCOD 2001 - Chilton, United Kingdom
Duration: Jul 9 2001Jul 11 2001

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2097
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference18th British National Conference on Databases, BNCOD 2001
Country/TerritoryUnited Kingdom
CityChilton
Period7/9/017/11/01

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'An analysis of main-memory and log space usage in extended ephemeral logging'. Together they form a unique fingerprint.

Cite this