Browse Prior Art Database

Distributed System Testbeds Experimentation with Distributed Sytems Guest Editor's Introduction Disclosure Number: IPCOM000131535D
Original Publication Date: 1982-Oct-01
Included in the Prior Art Database: 2005-Nov-11
Document File: 4 page(s) / 20K

Publishing Venue

Software Patent Institute

Related People

Helmut K. Berg: AUTHOR [+3]


Honeywell Corporate Computer Sciences Center

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 30% of the total text.

Page 1 of 4


This record contains textual material that is copyright ©; 1982 by the Institute of Electrical and Electronics Engineers, Inc. All rights reserved. Contact the IEEE Computer Society (714-821-8380) for copies of the complete work that was the source of this textual material and for all use beyond that as a record from the SPI Database.

Distributed System Testbeds Experimentation with Distributed Sytems Guest Editor's Introduction

Helmut K. Berg

Honeywell Corporate Computer Sciences Center

Distributed processing offers a potential for achieving system attributes such as extensibility, integrity, and performance. However, after years of research into specific issues of distributed computing and development of distributed systems, we still don't know whether this potential can be exploited to achieve the envisioned benefits. Many inherent problems need to be overcome before the benefits of distributed systems can be realized. In fact, most of what we know about these systems is theoretical, and few of these theories have been tested in actual distributed systems. Initial designs and developments of distributed systems indicate that extensibility, integrity, and performance may be poorer in distributed systems than in centralized ones. Among the causes of difficulties with distributed processing are a lack of experience with and data about distributed systems and a lack of appropriate methodologies and tools for designing distributed systems and their applications.

In solving these problems, we have been moving more and more towards experimental studies, which offer a

number of benefits. We expect experimentation with distributed systems to provide

an improved understanding of the functional requirements and operational behavior of distributed systems,

measurements from which quantitative results about distributed systems can be derived,

an integrated environment in which the interrelations of solutions to individual problems can be evaluated, and

a design environment in which design decisions can be based on both theoretical and empirical studies.

The move towards empirical investigations involves many areas of computer science, not just distributed processing. As Howden says,l "It is the author's opinion that the greatest practical benefits are to be gained from continued empirical rather than theoretical studies."

Scalf et al., in a report on the First International Conference on Distributed Computing, underlined the par ticular need for experimentation in distributed processin' when they wrote

Computer Science is a very pragmatic discipline. There are limits to the progress that can be achieved without experimental investigation. Direct experimental investigation in distributed computing requires one to invest in a number of processing elements and the requisite switching technology to provide the interconnections without the advantages of micr...