Bilkent University
Department of Computer Engineering


Resource Allocation and Provisioning in Edge Cloudlets


Dr. Onur Acgil
Postdoctoral Research Fellow, University College London

In traditional cloud computing model, intensive data processing required by applications takes place on resource-rich datacenters with powerful, co-located servers. Today this model is increasingly being challenged by the new and emerging applications that produce enormous amounts of datafor example, from smart/autonomous vehicles, wearables, sensors, and various other IoT devices (e.g., camera feeds), among othersdue to not only the network overheads involved in moving large volumes of data to centrally-located datacenters, but also the stringent latency requirements of applications. Instead, in the recently proposed edge computing model, computation can be moved closer to where the data is produced (i.e., at the "edges" of the network) and carried out at distributed computation spots (Cloudlets), each with very limited resource capacity. In this talk, I will present several resource allocation and provisioning algorithms for deployment of processing functions in the form of lightweight Virtual Machines (e.g., containers) at distributed computation spots for applications with strict low-latency requirements. In the last part of the talk, I will discuss several on-going work and future research directions.

Bio: Onur Ascigil is a Postdoctoral Research Fellow in the Electronic and Electrical Engineering Department at University College London. He received his PhD degree from Computer Science Department, University of Kentucky in 2014. His research interests are broadly in the area of networking and systems with a special focus on information-centric networking, edge-/fog-computing, and Internet-of-Things.


DATE: 09 January 2020, Thursday @ 13:40