Moa — a fast sliding compaction scheme for a large storage space

Mitsugu Suzuki, Hiroshi Koide, Motoaki Terashima

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

Abstract

The design and analysis of a new GC scheme called MOA is presented with its implementation on PLisp (Portable Lisp). MOA is "stop-and-collect" type GC and is based on a Morris's sliding compaction scheme. MOA has the excellent features such as: (1) it can perform sliding compaction with a time proportional nearly to the size of all data objects in use, (2) it requires an additional space of a small size to achieve such a time cost saving, (3) it can skip a GC process for a special cluster called an "anchor", reducing the total GC processing time considerably. MOA has been successfully implemented on PLisp which provides a large amount of storage space. MOA is superior to other GC based on conventional sliding compaction and copying collection, as shown in several experiments.

Original languageEnglish
Title of host publicationMemory Management - International Workshop IWMM 1995, Proceedings
EditorsHenry G. Baker
PublisherSpringer Verlag
Pages197-210
Number of pages14
ISBN (Print)3540603689, 9783540603689
DOIs
Publication statusPublished - Jan 1 1995
EventInternational Workshop on Memory Management, IWMM 1995 - Kinross, United Kingdom
Duration: Sep 27 1995Sep 29 1995

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume986
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherInternational Workshop on Memory Management, IWMM 1995
CountryUnited Kingdom
CityKinross
Period9/27/959/29/95

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Moa — a fast sliding compaction scheme for a large storage space'. Together they form a unique fingerprint.

Cite this