A framework for and implementation of intelligent task distribution

Abstract

There has been a large increase in the use of distributed systems to solve computationally expensive problems. There are many technologies available to researchers wishing to utilise a cluster of computers, however they often require the installation of software onto the hard disks of the machines being used and this can cause problems. There exist instances where, for one reason or another, the hard disks of the machines cannot be used. In this work we present detail of a framework and implementation of a distributed network that boots off a floppy disk and does not use the hard disks at all, but a RAMDrive. We go on to present details of how the server aspect of the framework can be used to effectively reduce load imbalance without the need for explicit user intervention.