Routing and scheduling in future data center networks

Jia, Zequn

Routing and scheduling in future data center networks

Jia, Zequn

Permalink

Publication Type:: Thesis
Issue Date:: 2024

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download thesisAdobe PDF (1.68 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Jia, Zequn
dc.date.accessioned	2025-05-23T01:27:45Z
dc.date.available	2025-05-23T01:27:45Z
dc.date.issued	2024
dc.identifier.uri	http://hdl.handle.net/10453/187474
dc.description	University of Technology Sydney. Faculty of Engineering and Information Technology.	en_US.UTF-8
dc.description.abstract	This thesis addresses critical challenges in software-defined data center networks by proposing solutions for routing control, fault detection, and traffic scheduling. The research introduces two routing methods: cRetor, which uses topology description language for efficient network setup, and sRetor, which distributes decision-making to switches to reduce controller load. Both methods demonstrate improved network performance through simulation results. For fault detection, the thesis presents two approaches: a heuristic algorithm and a deep reinforcement learning-based algorithm for generating fault probing matrices. These algorithms optimize detection paths and enhance efficiency and accuracy in network fault detection. Additionally, the research proposes a fault-aware traffic scheduling algorithm combining graph neural networks with deep reinforcement learning, which adapts to dynamic topology changes during network failures. This integrated approach demonstrates superior performance compared to existing methods. Overall, the thesis provides a comprehensive framework for intelligent adaptive network control, contributing to the advancement of software-defined data center networks' efficiency, reliability, and scalability.	en_US.UTF-8
dc.format	Thesis (PhD)
dc.language.iso	en_US	en_US.UTF-8
dc.relation	https://opus.lib.uts.edu.au/bitstream/10453/187474/1/thesis.pdf
dc.rights	info:eu-repo/semantics/openAccess
dc.rights	The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
dc.rights	© 2024 Zequn Jia
dc.rights	au.edu.uts.lib/nph
dc.title	Routing and scheduling in future data center networks	en_US.UTF-8
dc.type	Thesis
utslib.copyright.status	open_access	*

Abstract:

This thesis addresses critical challenges in software-defined data center networks by proposing solutions for routing control, fault detection, and traffic scheduling. The research introduces two routing methods: cRetor, which uses topology description language for efficient network setup, and sRetor, which distributes decision-making to switches to reduce controller load. Both methods demonstrate improved network performance through simulation results. For fault detection, the thesis presents two approaches: a heuristic algorithm and a deep reinforcement learning-based algorithm for generating fault probing matrices. These algorithms optimize detection paths and enhance efficiency and accuracy in network fault detection. Additionally, the research proposes a fault-aware traffic scheduling algorithm combining graph neural networks with deep reinforcement learning, which adapts to dynamic topology changes during network failures. This integrated approach demonstrates superior performance compared to existing methods. Overall, the thesis provides a comprehensive framework for intelligent adaptive network control, contributing to the advancement of software-defined data center networks' efficiency, reliability, and scalability.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/187474