Probabilistic-QoS-Aware Multi-Workflow Scheduling Upon the Edge Computing Resources

Abstract

Edge computing is an evolving decentralized computing infrastructure by which end applications are situated near the computing facilities. While the edge servers leverage the close proximity to the end-users for provisioning services at reduced latency and lower energy costs, their capabilities are constrained by limitations in computational and radio resources, which calls for smart, quality-of-service (QoS) guaranteed, and efficient task scheduling methods and algorithms. For addressing the edge-environment-oriented multi-workflow scheduling problem, the authors consider a probabilistic-QoS-aware approach to multi-workflow scheduling upon edge servers and resources. It leverages a probability-mass function-based QoS aggregation model and a discrete firefly algorithm for generating the multi-workflow scheduling plans. This research conducted an experimental case study based on varying types of workflow process models and a real-world dataset for edge server positions. It can be observed the method clearly outperforms its peers in terms of workflow completion time, cost, and deadline violation rate.