Performance Evaluation of a Parallel Knowledge-Based System

From Tetherless World Wiki

Jump to: navigation, search

Citation: D. Muliawan. (1989) Performance Evaluation of a Parallel Knowledge-Based System. In KSL-89-51, June,1989.

Publication techreport ( Edit )
type Technical Report
bibtype techreport
Bibtex basics
author D. Muliawan
title Performance Evaluation of a Parallel Knowledge-Based System
number KSL-89-51
institution Knowledge Systems, AI Laboratory
year 1989
month June
Bibtex more
Access Paper
abstract This paper discusses the quantitative and qualitative performance of a module of a parallel knowledge-based system for tracking and classifying aircraft, called Airtrac. The Airtrac system is built to gain some understanding of the potential speed up through concurrency of reasonably large and complex continuous signal understanding systems. Airtrac runs on CARE, a simulated distributed-memory, asynchronous message-passing multicomputer. Evaluating the performance of a continuous parallel knowledge-based system such as Airtrac is difficult. The simple approach of timing its execution would not work, since the system is continuous. Furthermore, the performance is usually multidimensional and cannot easily be expressed into a single number. The notions of latency, excess ratio, sustainable data rate, and capacity are instead used to rate the performance of the system. The paper reports the effects of important high-level control strategies on the system performance, the effects of varying the frequency and width of the input data across different numbers of processors, and some possible speed up curves of the overall system performance as a function of the number of processors. Finally conclusions are presented in the relationship between the quantitative and qualitative performance of Airtrac, and in the potential speedup of large and complex parallel knowledge-based systems.

KSL Technical Report ID: KSL-89-51
Facts about Performance Evaluation of a Parallel Knowledge-Based SystemRDF feed
Abstract This paper discusses the quantitative and … This paper discusses the quantitative and qualitative performance of a module of a parallel knowledge-based system for tracking and classifying aircraft, called Airtrac. The Airtrac system is built to gain some understanding of the potential speed up through concurrency of reasonably large and complex continuous signal understanding systems. Airtrac runs on CARE, a simulated distributed-memory, asynchronous message-passing multicomputer. Evaluating the performance of a continuous parallel knowledge-based system such as Airtrac is difficult. The simple approach of timing its execution would not work, since the system is continuous. Furthermore, the performance is usually multidimensional and cannot easily be expressed into a single number. The notions of latency, excess ratio, sustainable data rate, and capacity are instead used to rate the performance of the system. The paper reports the effects of important high-level control strategies on the system performance, the effects of varying the frequency and width of the input data across different numbers of processors, and some possible speed up curves of the overall system performance as a function of the number of processors. Finally conclusions are presented in the relationship between the quantitative and qualitative performance of Airtrac, and in the potential speedup of large and complex parallel knowledge-based systems. complex parallel knowledge-based systems.
Author D. Muliawan  +
Bibtype techreport  +
Has author D. Muliawan  +
Has identifier KSL-89-51  +
Has publishing details June,1989  +
Has title Performance Evaluation of a Parallel Knowledge-Based System  +
Has where published KSL-89-51  +
Has year 1989  +
Institution Knowledge Systems, AI Laboratory  +
Ksl tr id KSL-89-51  +
Month June  +
Number KSL-89-51  +
Process note NO  +
Title Performance Evaluation of a Parallel Knowledge-Based System  +
Year 1989  +
Retrieved from "http://tw.rpi.edu/wiki/Performance_Evaluation_of_a_Parallel_Knowledge-Based_System"
Personal tools