A migration tool to support resource and load sharing in heterogeneous computing environments

  • Authors:

  • P.D Le;B Srinivasan

  • Affiliations:

  • Department of Computer Technology, Monash University, Australia;Department of Computer Technology, Monash University, Australia

  • Venue:
  • Computer Communications
  • Year:
  • 1997

Quantified Score

Hi-index 0.24

Visualization

Abstract

Distributed computing systems potentially provide significant advantages and benefits such as enhanced performance, extensibility, reliability, and better resource and load sharing. However, the evolution of hardware and software has unavoidably led to heterogeneous computing systems. The heterogeneity of such systems gives rise to many problems such as transparency, application developments, database and network management, efficient resource and load sharing, migration, etc. This paper discusses the problems associated with resource and load sharing in such heterogeneous environments. First, it discusses different techniques such as distributed file systems, network operating systems, load-sharing algorithms, migration mechanisms and other methods which have been proposed as solutions for resource and load sharing. Then it describes our design and implementation of a prototype tool for program and data migration in heterogeneous computing environments of different computer architectures and different operating systems. The tool was based on the client-server model and the use of remote procedure calls for interconnecting heterogeneous distributed computer systems. This important decision supports the evolution of the computer hardware and software without side-effects. It adopts well with the evolution of distributed systems as the degree of heterogeneity increases. The tool aims to allow better resource and load sharing in heterogeneous computer systems; to allow useful information to be quickly obtained; to reduce communication costs and programmers' time in prototyping distributed applications. It is designed and implemented on a local network of computers connected by Ethernet and running different Unix and Unix-like operating systems.