01whole.pdf (44.53 MB)
Download file

Lazy max weight scheduling

Download (44.53 MB)
thesis
posted on 28.03.2022, 20:02 by Christelle Haddad
Scheduling, resource allocation and load balancing applications are present in all walks of life. Whether it be waiting at a set of traffic lights, in line at a cash register, or in a communication network, these everyday tasks have one thing in common. However, one aspect of resource allocation which has yet to be optimised, is when switch over delays are imposed Such delays are very common in systems where there must be a reconfiguration of the system's functionality. Some examples include satellite and optical communications, or more simply, the changing of traffic lights at an intersection. Over time, many potential solutions have been proposed on how to minimise the effects of such switch over delays, but these efforts are yet to provide a single, optimal solution. This said, this paper will investigate the behaviour of simulations implementing the theoretical and mathematical insights into the functionality of the Variable Frame-based Max Weight (VFMW) resource allocating policy. As implied by the name of the VFMW policy - frames of various sizes are allocated to particular service vectors in an attempt to mitigate the effects of the reconfiguration delays of which the server is idle. Therefore, a number of additional frame determining functions will be investigated, allowing a potential solution to minimising the system delay while ensuring the system achieves a steady state result. An additional extension to this VFMW policy will also be presented through the simulation to provide evidence that the inclusion of unnecessary reconfiguration delays can have greatly detrimental effects on systems of particular traffic loads.

History

Table of Contents

1. Introduction -- 2. Theory -- 3. System model -- 4. Simulation outline -- 5. Simulation results -- 6. Analysis -- 7. Conclusion -- 8. Abbreviations and variables -- Appendices -- Bibliography.

Notes

Bibliography: pages 215-216 Empirical thesis.

Awarding Institution

Macquarie University

Degree Type

Thesis bachelor honours

Degree

BSc (Hons), Macquarie University, Faculty of Science and Engineering, School of Engineering

Department, Centre or School

School of Engineering

Year of Award

2016

Principal Supervisor

Phillip Whiting

Rights

Copyright Christelle Haddad 2016. Copyright disclaimer: http://mq.edu.au/library/copyright

Language

English

Extent

1 online resource (xii, 216 pages graphs, tables)

Former Identifiers

mq:70320 http://hdl.handle.net/1959.14/1262525

Usage metrics

Keywords

Exports