A queueing network model for the effect of data compression on system efficiency

Authors:
Alan Jay Smith
Affiliations:
University of California at Berkeley, Berkeley, California
Venue:
AFIPS '76 Proceedings of the June 7-10, 1976, national computer conference and exposition
Year:
1976

Citing 3
Cited 0

Open, Closed, and Mixed Networks of Queues with Different Classes of Customers

Journal of the ACM (JACM)
An Introduction to Database Systems

An Introduction to Database Systems
A performance analysis of multiple channel controllers

SIGME '73 Proceedings of the 1973 ACM SIGME symposium

Quantified Score

Hi-index	0.00

Visualization

Abstract

Data compression is often employed in data base systems to reduce the volume and therefore the cost of large online data bases. The use of data compression is normally assumed to imply a tradeoff---storage space is decreased in return for which the CPU is required to spend time coding and decoding the data, with a consequent decline in system throughput. We employ queueing network models with classes of customers to show that the effect of data compression on system efficiency can be calculated. Our calculations indicate that under some circumstances, data compression may actually increase system throughput by decreasing input/output transfer time and/or increasing the useful processing per data record. We formulate and solve several models to indicate the power of our technique.