Spatiotemporal Non-Uniformity-Aware Online Task Scheduling in Collaborative Edge Computing for Industrial Internet of Things

TL;DR

The paper tackles online task scheduling in collaborative edge computing for IIoT under long-term cost budgets, addressing both spatial non-uniformity across factories and time-varying request distributions. It develops a Lyapunov-based framework that transforms the long-horizon objective into per-slot problems, augmented by a graph model that captures the spatiotemporal structure of requests and resources. A two-stage heuristic (enhanced discrete particle swarm plus harmony search) solves the per-slot problem, and an imitation-learning module accelerates operation by predicting category assignments to reduce search. Theoretical analysis provides delay-cost bounds and complexity results, and extensive simulations plus a Kubernetes-based prototype validate improved latency and budget adherence compared to baselines, confirming practical applicability. The approach enables scalable, real-time scheduling for industrial edge ecosystems with privacy-preserving, horizontal collaboration across factories.

Abstract

Mobile edge computing mitigates the shortcomings of cloud computing caused by unpredictable wide-area network latency and serves as a critical enabling technology for the Industrial Internet of Things (IIoT). Unlike cloud computing, mobile edge networks offer limited and distributed computing resources. As a result, collaborative edge computing emerges as a promising technology that enhances edge networks' service capabilities by integrating computational resources across edge nodes. This paper investigates the task scheduling problem in collaborative edge computing for IIoT, aiming to optimize task processing performance under long-term cost constraints. We propose an online task scheduling algorithm to cope with the spatiotemporal non-uniformity of user request distribution in distributed edge networks. For the spatial non-uniformity of user requests across different factories, we introduce a graph model to guide optimal task scheduling decisions. For the time-varying nature of user request distribution and long-term cost constraints, we apply Lyapunov optimization to decompose the long-term optimization problem into a series of real-time subproblems that do not require prior knowledge of future system states. Given the NP-hard nature of the subproblems, we design a heuristic-based hierarchical optimization approach incorporating enhanced discrete particle swarm and harmonic search algorithms. Finally, an imitation learning-based approach is devised to further accelerate the algorithm's operation, building upon the initial two algorithms. Comprehensive theoretical analysis and experimental evaluation demonstrate the effectiveness of the proposed schemes.

Figures (15)

• Figure 1: Illustration of the network architecture for collaborative edge computing in the IIoT
• Figure 2: Computing network graph for each service corresponding to Fig. \ref{['Figure:System Model']} during the $t$th time slot. (a) Computing network graph of app_1. (b) Computing network graph of app_2. (c) Computing network graph of app_3. (d) Computing network graph of app_4. Solid and hollow circles represent cases where $F_{m}^{k}(t)>0$ and $F_{m}^{k}(t)=0$, respectively.
• Figure 3: Illustration of the processing flow for requests against the $k$th service initiated by IIoT devices in Factory 1
• Figure 4: An example of a weighted directed graph for the task scheduling decision of the $k$th service at the $t$th time slot. This example shows that, in the current time slot, the AP of factory 2 will send $N_{2,3}^k(t)$ requests for $app\_k$ to the ES of factory 3 and $N_{2,4}^k(t)$ requests to the ES of factory 4. The AP of factory 1 will send $N_{1,3}^k(t)$ requests for $app\_k$ to the ES of factory 3.
• Figure 5: Online task scheduling framework.

Theorems & Definitions (11)

• Theorem 1
• proof
• Theorem 2
• proof
• Theorem 3
• proof
• Theorem 4
• proof
• Corollary 1
• Theorem 5