Independent task scheduling algorithms in fog environments from users’ and service providers’ perspectives: a systematic review

Fog computing has been touted as a solution for the massive growth in the need for processing Internet of Things (IoT) tasks efficiently by bringing resources closer to the edge of network, reducing the need to transmit tasks and processing to the cloud. However, new challenges such as limited computation, high levels of complexity and the distributed nature of fog devices arise, which affect the system’s quality of service (QoS). To address these issues, a substantial amount of work has been done to determine optimal methods to improve QoS using task scheduling algorithms. This review classifies the scheduling algorithms according to their QoS impact on both users and fog service providers (FSPs). It is important to consider a balance of objectives and QoS parameters for both FSPs and users to meet requirements of all stakeholders. This study extensively analyses articles based on objectives, scheduling techniques, simulation tools, and computing environments. To assure unbiased evaluation, a benchmark to assess task scheduling algorithms is proposed based on comparison to state-or-art algorithms, applicability to real-world contexts, meeting the objectives of the study, negative impacts on other performance metrics, and sizes of datasets used for evaluation. Of the 66 articles surveyed, 22 met all evaluation criteria in the different fog environments and objectives, with 16 meeting both FSPs and users' QoS requirements. Additionally, a number of gaps for future research were uncovered, including the lack of research on scheduling for IoT-fog-cloud and IoT-fog environments, and the lack of investigation considering conflicting objectives such as deadline violation-reliability, response time-energy consumption, and so on.

Text 121688.pdf - Published Version Restricted to Repository staff only Download (2MB) Official URL or Download Paper: https://link.springer.com/article/10.1007/s10586-0...

Actions (login required)

View Item