A Dynamic Load Balancing Architecture for PDES Using PVM on Clusters

Authors:
Arnold N. Pears;Nicola Thong
Affiliations:
-;-
Venue:
Proceedings of the 8th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Year:
2001

Citing 6
Cited 1

Distributed discrete-event simulation

ACM Computing Surveys (CSUR)
A survey of discrete event simulation and state-of-the-art discrete event languages

ACM SIGSIM Simulation Digest
Parallel discrete event simulation

Communications of the ACM - Special issue on simulation
PVM: Parallel virtual machine: a users' guide and tutorial for networked parallel computing

PVM: Parallel virtual machine: a users' guide and tutorial for networked parallel computing
Trade-Off between Sequential and Time Warp-Based Parallel Simulation

IEEE Transactions on Parallel and Distributed Systems
Parallel simulation using the time warp operating system (tutorial session)

WSC' 90 Proceedings of the 22nd conference on Winter simulation

Explicit Spatial Scattering for Load Balancing in Conservatively Synchronized Parallel Discrete Event Simulations

PADS '10 Proceedings of the 2010 IEEE Workshop on Principles of Advanced and Distributed Simulation

Quantified Score

Hi-index	0.00

Visualization

Abstract

Scalable simulation in a cluster computing environment depends on effective mechanisms for load balancing. This paper presents an object-oriented software architecture and specifies communication protocols which support efficient load balancing via object migration in cluster based PDES. A general decomposition of functionality is given in the context of a flexible adaptable architecture for migration and load adjustment. The key contributions are the development of a general architecture into which specific migration policies and load balancing heuristics can be easily integrated and a specification of the communication protocols required. An implementation of the proposed architecture using PVM for communication indicates the approachis both flexible and well suited to customization. This enhances simulation scalability and usability in computing platforms including both traditional cluster computing and Web-computing.