About the Project
Background: Monitoring data-center status plays a critical role in Cloud data-center management: it provides the intelligence for data-center optimization. The problem is that distributed state monitoring is hard in large-scale dynamic data-centers. Existing periodic monitoring delivers bad intelligence, as it is very costly and not very accurate. The objective of this PhD is to use Machine Learning to perform distributed event-based, adaptive monitoring, that delivers faster, more accurate intelligence, using Software Defined Networking.
This PhD builds upon the research team’s recent results in the area of (1) predicting Quality of Delivery (QoD) metrics for adaptive video codec sessions in SDNs; (2) rapid restoration techniques for software-defined networks; and (3) video QoD prediction under time-varying loads in data-centre networks. Improving monitoring will significantly improve the intelligence used for path reallocation and data-center performance in these different application domains.
This PhD research project is part of a Science Foundation Ireland SIRG project, which sits at the intersection of Computer Networks, Machine Learning and Signal Processing, areas of intense interest for the global research and industrial community at present. The team has worked on developing and analyzing algorithms for scalable distributed network monitoring and service-level prediction in the Cloud, in collaboration with leading academic and industry partners, e.g. KTH Royal Institute of Technology (Stockholm), University of Portsmouth, IBM (Dublin), Amadeus SAS (Sophia Antipolis), etc. The successful candidate will be encouraged to form research collaborations by participating in national and international conferences.
Research Environment: The successful applicant will work with an existing world-class team in this area in TU Dublin, in the School of Electrical and Electronic Engineering. The PhD will benefit from the ongoing research in the SFI CONNECT Research Center (https://connectcentre.ie/) as well as the SFI Advance (https://www.advance-crt.ie/) CRT programme.
The successful application will develop new monitoring techniques by performing theoretical analysis and then by designing and testing the resulting protocols in different networking scenarios. Some example outputs of this work will include:
- (1) Definition of learning models for monitoring scenarios;
(2) Creation of a SDN frameworks to evaluate new monitoring protocols (mininet, various SDN controllers, D-ITG, etc.)
- (3) Development of machine learning algorithms for calculating when and how to monitor also with comparison with the SOTA. Study of the analytical properties of the strategies. Consideration of multi-task applications.
(4) Performing performance evaluations and drawing conclusions on which monitoring strategy is the most beneficial in different scenarios.
Student requirements for this project
Eligibility: Applicants should have (or expect to attain, subject to final examination results) a First or Upper Second Class Honours Degree in Electronic Engineering, Computer Science, Mathematics, or a related discipline.
Knowledge & Experience: The successful candidate should be self-motivated with the enthusiasm to develop technical skills across a range of disciplines, including Signal Processing, Network Emulation, Machine Learning and Computer Networking. The following experience is desirable:
– Experience with Software Defined Networking;
– Experience of working on large datasets;
– Experience of applying Machine Learning/Signal Processing techniques to real-world data; – Experience of the evaluation of networking or management systems;
Skills & Competencies
– Strong analytical/mathematical skills;
– Applicants whose first language is not English must submit evidence of competency in English;
– Strong computer programming skills with a language such as Python, C++ etc.
If you are interested in submitting an application for this project, submit an electronic copy of your Curriculum Vitae, your transcripts, proof of English competency scores (IELTS), and a letter of motivation (which outlines why you have chosen to apply, why you have chosen this supervision team and what you think the project will be about) with the names and email addresses of two references to [Email Address Removed].
Materials/ Travel etc NA
This project is part of the Science Foundation Ireland Starting Investigator Research Grant (SIRG). Material/Travels are covered by the project budget. It is not specified on a per student basis.
Based on your current searches we recommend the following search filters.
Based on your current search criteria we thought you might be interested in these.