Two Classical Scheduling Strategies in Application to Distributed Computing

📄 Project Abstract

Scheduling tasks on a distributed system, while minimising the time at which the final task completes, is an interesting but difficult, well studied problem in Computer Science. Two communities have developed many heuristics in an attempt to solve this problem in a variety of forms; one has been subject to traditional classical Scheduling research for over 50 years, the other, whose research surrounds distributed systems, is a relatively new area of study. In classical Scheduling theory, it is shown that a particular algorithm is preferred over a similar but opposite rule. In Distributed Computing, it is shown that an equivalent algorithm to this opposite rule is produces better results. We show the equivalence of the corresponding algoritms within the two communities. Following that we produce software and perform experiments with this software in an attempt to satisfy the five hypotheses that are proposed. Using the results of the experiments we explain the contradiction described ab...