Skip to main navigation Skip to search Skip to main content

Parallel algorithms for maximal acyclic sets

  • Tokyo Denki University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Given a graph G=(V, E), the classical spanning forest problem of G can be viewed as the problem of finding a maximal subset F of E inducing an acyclic subgraph. Although it is well known that this problem has efficient NC algorithms, its vertex counterpart, i.e., the problem of finding a maximal subset U of V inducing an acyclic subgraph, has not been shown to be in NC (or even in RNC) and is not believed to be parallelizable in general. We present NC algorithms for solving the latter problem for three special cases. The first algorithm solves the problem for planar graphs in O(log3 n) time using O(n) processors on an EREW PRAM. The second algorithm solves the problem for K3,3-free graphs in O(log4 n) time using O(n) processors on an EREW PRAM. The third algorithm solves the problem for graphs without long induced paths in poly-logarithmic time using O(n2376) processors on an EREW PRAM.

Original languageEnglish
Title of host publicationProceedings - 1st Aizu International Symposium on Parallel Algorithms/Architecture Synthesis, AISPAS 1995
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages169-175
Number of pages7
ISBN (Electronic)081867038X, 9780818670381
DOIs
StatePublished - 1995
Event1st Aizu International Symposium on Parallel Algorithms/Architecture Synthesis, AISPAS 1995 - Aizu-Wakamatsu, Fukushima, Japan
Duration: Mar 15 1995Mar 17 1995

Publication series

NameProceedings - 1st Aizu International Symposium on Parallel Algorithms/Architecture Synthesis, AISPAS 1995

Conference

Conference1st Aizu International Symposium on Parallel Algorithms/Architecture Synthesis, AISPAS 1995
Country/TerritoryJapan
CityAizu-Wakamatsu, Fukushima
Period03/15/9503/17/95

Fingerprint

Dive into the research topics of 'Parallel algorithms for maximal acyclic sets'. Together they form a unique fingerprint.

Cite this