NUMBER = "TI-791" INTERNAL_NUMBER = "THD-AT-1991-07" 
TITLE = "The Learning Behaviour of a Scheduler using a Stochastic Learning
Automaton" 
AUTHOR = "Thomas Kunz" 
ORGANIZATION = "University of Darmstadt" 
DEPARTMENT = "Dept. of Computer Science (FB20)"
INSTITUTE = "Institute for Theoretical Computer Science" 
DIVISION
= "Automata Theory and Formal Languages Division" 
DATE ="December 1991"

ABSTRACT 

This paper discusses a load balancing heuristic in a
general-purpose distributed computer system. We implemented a
task scheduler based on the concept of a Stochastic Learning
Automaton on a network of Unix workstations. The used heuristic
and our implementation are shortly  described. Creating an
executable artificial workload, a number of  experiments examined
different learning schemes. Using a linear  reward--penalty
scheme resulted in the best performance of the scheduler. Another
series of experiments looked at different ways to evaluate the
goodness of a scheduling decision, another aspect of the learning
behaviour. Instead of using a simple binary (qualitative)
measure, a quantitative evaluation allowed for a more stable and
therefore better learning behaviour." KEYWORDS  "Distributed
systems, heuristics, load balancing, queueing systems, stochastic
learning automata"