Browse Prior Art Database

Issues and Approaches to Distributed Testbed Instrumentation

IP.com Disclosure Number: IPCOM000131540D
Original Publication Date: 1982-Oct-01
Included in the Prior Art Database: 2005-Nov-11
Document File: 16 page(s) / 54K

Publishing Venue

Software Patent Institute

Related People

William R. Franta: AUTHOR [+5]

Abstract

Centralized control is fairly common to distributed system testbeds, although its form varies for,different systems. This testbed extends the concept and takes an extreme approach. Computer system development requires a variety of modeling and analysis techniques. These include queueing networks for mathematical analysis, Petri nets for concurrency descriptions, state transition assertions for program verification, and simulation models for investigations of operational behavior. When theoretical or simulation postulations appear inappropriate, hardwarebased experimental studies are often used. Hardwarebased experiments for computer system development focus on assessing system and component behavior, and experimental investigations require an environment that facilitates the observation of that behavior. Experience gained in the development of experimentation environments for centralized computer systems has been used to establish methodologies and tools for centralized system experimentation. For distributed computer systems, however, there is not yet a rich experience base for developing tools and methodologies to provide an experimental environment.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 7% of the total text.

Page 1 of 16

THIS DOCUMENT IS AN APPROXIMATE REPRESENTATION OF THE ORIGINAL.

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 http://www.computer.org/ (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.

Issues and Approaches to Distributed Testbed Instrumentation

William R. Franta University of Minnesota Helmut K. Berg and William T. Wood Honevwell Corporation

Centralized control is fairly common to distributed system testbeds, although its form varies for,different systems. This testbed extends the concept and takes an extreme approach.

Computer system development requires a variety of modeling and analysis techniques. These include queueing networks for mathematical analysis, Petri nets for concurrency descriptions, state transition assertions for program verification, and simulation models for investigations of operational behavior. When theoretical or simulation postulations appear inappropriate, hardwarebased experimental studies are often used. Hardwarebased experiments for computer system development focus on assessing system and component behavior, and experimental investigations require an environment that facilitates the observation of that behavior.

Experience gained in the development of experimentation environments for centralized computer systems has been used to establish methodologies and tools for centralized system experimentation. For distributed computer systems, however, there is not yet a rich experience base for developing tools and methodologies to provide an experimental environment.

A distributed system testbed -- a vehicle for experimentation -- is distinguished from a distributed system by facilities that allow the distributed system topology to be varied and the behavior of the distributed system to be readily observed and recorded. This difference gives rise to the logical structure of a distributed system testbed. It generally contains three major components: a multinode experimentation subsystem, an instrumentation subsystem, and a support subsystem.

Each node in the multinode experimentation subsystem contains a computer with an autonomous control nucleus, interfaces to the communication network, and input/output facilities. The physically separated experimentation nodes retain processes and resources that are fragments of overall system processing activity. A node does not need access to the (disjoint) memory spaces of other nodes, nor does it need systemwide control authority. Rather, system control functions and state in formation are generally partitioned and distributed among the nodes, with system control provided through concurrent processes that interact via message-based communication.

The instrumentation subsystem monitors the operation of the experimentation subsystem...