Bilateral Preference Algorithm Optimizes Task Scheduling and Energy in Edge Computing

A Bilateral Preference-Aware Task Scheduling Algorithm, designated as BPA, has been proposed for edge computing environments to improve overall system utility while balancing task latency and energy consumption. Within edge computing architectures, efficient task scheduling represents a critical operational requirement for enhancing system performance and utility. Managing tasks effectively is necessary to ensure that computational workloads do not cause excessive latency while simultaneously keeping energy expenditure under control. The newly developed BPA method is designed to address this challenge directly by establishing a scheduling framework that incorporates mutual preferences between incoming Internet of Things tasks and edge servers.

The proposed algorithm models mutual preferences between computational tasks and edge servers to capture both task requirements and resource characteristics. In this formulation, tasks possess distinct requirements that must be accommodated, while edge servers have specific resource characteristics that govern how they can process assigned workloads. By explicitly representing these mutual preferences, the algorithm ensures that both task-side demands and server-side capabilities are taken into account during scheduling.

To make effective use of these preferences, BPA applies a structured two-stage optimization strategy to guide its scheduling decisions. In the first optimization stage, the algorithm adjusts processor frequency through dynamic voltage and frequency scaling, widely referred to as DVFS. In the second optimization stage, the framework refines the scheduling results via task exchanges across the edge servers. By allowing tasks to be exchanged and reassigned among edge servers, this second phase refines the preliminary allocation to further enhance overall system utility.

Extensive simulation results demonstrate that, compared with traditional scheduling algorithms, the proposed BPA approach significantly improves the utilization of system resources. The simulations evaluated the algorithm across diverse load conditions, demonstrating that it consistently delivers strong overall performance. Throughout these varied conditions, the framework achieves a well-balanced trade-off between task latency and energy consumption. By coordinating dynamic voltage and frequency scaling with subsequent task exchanges, BPA provides an effective and balanced mechanism for managing task execution and server resources across edge computing environments.

Subject of Research: IoT task scheduling in edge computing

Article Title: Bilateral preference-aware IoT task scheduling approach for edge computing

Article References: Hou, S., Xu, Y., Zou, W., Tong, Q., Yu, A., & Liu, X. (2026). Bilateral preference-aware IoT task scheduling approach for edge computing. Cluster Computing, 29(16), Article 931. https://doi.org/10.1007/s10586-026-06712-7

Image Credits: AI Generated

DOI: 10.1007/s10586-026-06712-7

Keywords: edge computing, task scheduling, Internet of Things, energy consumption, task latency, resource utilization, dynamic voltage and frequency scaling, optimization algorithms, Bilateral, preference-aware, task, scheduling