Task partitioning and orchestration in edge computing systems

Abstract

Many computer-intensive applications running on mobile devices require low latency responses, such as 3D SLAM. Edge computing is a new paradigm moving Cloud features closer to end users and providing necessary computing and network resources at the edge of the network. In this talk, I will present the EDGEVISION framework for computer vision applications partitioning and orchestration on heterogeneous edge computing platforms considering both CPUs and GPUs. EDGEVISION abstracts the heterogeneous hardware resources and the task runtime environments and divides the application into separate tasks to be orchestrated and deployed into the heterogeneous edge nodes. We also propose two scheduling algorithms in our framework, MLTS and MCTS, aiming to minimize the processing latency and the overall system cost. We evaluate our framework by implementing the edge-based 3D SLAM application in our real testbed with ten heterogeneous edge devices.