On Network Reliability Evaluation by Monte Carlo Method Using High-Performance Computing

The paper considers the NP-hard problem of calculation the reliability of a network, which elements are subject to accidental failures. As network reliability, we mean the probabilistic connectivity of a random graph with unreliable edges. To evaluate the reliability of a network, a parallel Monte C...

Full description

Saved in:
Bibliographic Details
Published inLobachevskii journal of mathematics Vol. 44; no. 8; pp. 3122 - 3129
Main Authors Migov, D. A., Weins, D. V.
Format Journal Article
LanguageEnglish
Published Moscow Pleiades Publishing 01.08.2023
Springer Nature B.V
Subjects
Algebra
Algorithms
Analysis
Geometry
Graph theory
Mathematical Logic and Foundations
Mathematics
Mathematics and Statistics
Monte Carlo simulation
Multiagent systems
Network reliability
Probability Theory and Stochastic Processes
Reliability analysis
System reliability
parallel algorithms
Monte Carlo methods
random graph
multi-agent simulation
connectivity
MPI
network reliability
Online AccessGet full text

Cover

More Information
Summary:The paper considers the NP-hard problem of calculation the reliability of a network, which elements are subject to accidental failures. As network reliability, we mean the probabilistic connectivity of a random graph with unreliable edges. To evaluate the reliability of a network, a parallel Monte Carlo method is used, improved by checking the connectivity of a particular graph realization simultaneously with the generation of this realization. Based of multi-agent simulation, we study the scalability of this algorithm and tune the parameters for an execution using high-performance supercomputers.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1995-0802
1818-9962
DOI:10.1134/S1995080223080413