英国Queen's University Belfast大学老师Blesson Varghese招CS博士生

ly5490UID: 3832319

PhD Opportunity – Project LEEWAY
PhD Opportunity at Queen’s University Belfast
LEEWAY – Lean Efficient Edge-based WorkloAd DeploYment on the Internet

The emerging Internet-of-Things (IoT) envisions connecting billions of people and things including mobile devices and sensors to the internet. To facilitate this nodes located at the edge of the network, referred to as edge nodes, will need to be leveraged for deploying latency sensitive workloads that service requests from mobile devices or sensors. Edge nodes may take the form of dedicated compute resources, such as a cluster of low power devices or traffic routing nodes, such as routers, switches and gateways, which offer their spare resources for computing. Currently, edge nodes are not included in the computing ecosystem and are not employed for general purpose computing.

It is challenging using current technologies to deploy workloads in a heterogeneous, multi-tenant and resource constrained environment that is typical of edge nodes. This is because current technologies that make use of virtualisation techniques for workload deployment, such as virtual machines or containers have a relatively large resource footprint and operation overhead. A radically new and alternate technology will need to be developed that can operate with limited overheads in resource constrained environments

This project will explore how workloads can be deployed in a lightweight and efficient manner across multiple hardware architectures. The project will develop by setting up a test-bed using cloud data center resources (typical of Intel Xeon servers or the like) and edge-based resources (such as Odroid Boards or Raspberry Pis) initially. A literature analysis on workload deployment techniques will be carried out to identify the gaps in current technologies. Then the primary goal of the project to design, develop and test cutting edge workload deployment techniques suited for edge-based IoT systems will be pursued.

Click on the following links for more background information:

The development of edge computing
https://arxiv.org/pdf/1707.07452.pdf

The challenges and opportunities in edge computing
https://arxiv.org/pdf/1609.01967.pdf

Resource management techniques at the edge of the network
http://bvarghese.staff.cs.qub.ac.uk/papers/ENORM-Varghese.pdf

The successful candidate will be supported in learning the fundamental concepts of edge computing systems and workload deployment techniques. The candidate will be integrated into an ambitious team with a mission to tackle challenges in developing next generation parallel and distributed computing systems. The project will develop with regular meetings with the supervisors. It is expected that the candidate will be dedicated and hardworking and will take a strong initiative in driving the project. The candidate will have the flexibility in defining the scope of this cutting edge project. Access to state-of-the-art hardware and software required for this research will be available. An exceptional range of training and mentoring tools for professional and personal development will be accessible. Target venues for publication include top IEEE and ACM journals and conferences. It is anticipated that the software deliverables from this project will be made publicly available or integrated into an open-source project.

Contact
Interested candidates are strongly encouraged to make contact with:
Dr Blesson Varghese (b.varghese@qub.ac.uk; http://www.blessonv.com), or
Dr Ivor Spence (i.spence@qub.ac.uk)