The focus of this research is on the study and development of sustainable (i.e.: energy efficient) solutions for Multimedia Event Processing over the Cloud and Edge, while accounting for the accuracy and latency of the user queries, according to the query-aware Quality of Service (QoS) scheduling policies. This solution will make use of a self-adaptive system to adapt to changes in the environment, user requirements, and system workload. In addition, Reinforcement Learning techniques and a Fuzzy Control system for uncertainty management and will be developed, to better cope with the intrinsic uncertainties in the adaptation life-cycle.
Felipe Arruda Pontes
Dr Edward Curry
Institution: NUI Galway
|||Felipe Arruda Pontes, Edward Curry, "Cloud-Edge Microservice Architecture for DNN-based Distributed Multimedia Event Processing", In 1st International Workshop on Edge Migration and Architecture (EdgeWays 2020), 2020. [bib] [pdf]|