Survey on Distributed Computing Networks - Networks of Workstations

By Bhavana Nagendra

Abstract: Distributed system is a programming infrastructure which allows the use of a collection of workstations as a single integrated system. Distributed Computing offers advantages for improving availability and reliability through replication, performance through parallelism and in addition flexibility, expansion and scalability of resources. This paper is a survey on the ongoing research in this area.

Table of Contents

  1. Introduction
  2. Distributed Computing Networks
    1. Changing Trends in Computing
    2. Distributed System Architecture
    3. Massively Parallel Processors (MPPs)
    4. Networks of Workstations (NOWs)
      1. Challenges for Networks of Workstations
      2. What is new about NOWs ?
      3. Opportunities for NOW
  3. Message passing Issues and Distributed System Testing Software
    1. Communication Paradigms
      1. Remote Procedure Call
      2. Distributed Shared Memory
    2. Software Packages
      1. Parallel Virtual Machine (PVM)
      2. The p4 System
      3. Express
      4. Message Passing Interface (MPI)
      5. The Linda System
  4. Advances in Networks of Workstations (NOWs)
    1. Low over-head communication
    2. GLUnix : A global layer Unix
    3. xFS : Serverless network file service
  5. Massive Parallelism in WANs
    1. Concepts
    2. Managing Parallelism in WANs
    3. Implementation Issue
  6. Conclusion
  7. Appendix - Annotated bibiliography
View the complete report online

Shift-click to download the report in Adobe Acrobat format (89,223 Bytes) |

Other reports in this series
Complete list of student reports on Recent Advances in Networking
Back to Raj Jain's Home Page