Graph Neural Network-Based Collaborative Perception for Adaptive Scheduling in Distributed Systems

2025-05-22

AI Summary

To address the challenges of fragmented multi-node perception and delayed scheduling responses in distributed systems, this paper constructs a graph-structured system model and proposes a perception representation method that jointly encodes local node states and global topological features. It further designs a multi-layer graph neural network (GNN)-driven adaptive collaborative scheduling framework, robust under bandwidth constraints and dynamic topology changes. The key innovation lies in the first integration of message-passing and state-update mechanisms within a unified graph model, enabling real-time inter-node cooperative perception and accurate global state inference. Experimental results demonstrate significant improvements over state-of-the-art algorithms across all major metrics: task completion rate, average latency, load balancing, and transmission efficiency. The proposed model exhibits rapid convergence and high responsiveness to complex, time-varying system dynamics.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Multiagent Systems: Distributed Problem Solving

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

This paper addresses the limitations of multi-node perception and delayed scheduling response in distributed systems by proposing a GNN-based multi-node collaborative perception mechanism. The system is modeled as a graph structure. Message-passing and state-update modules are introduced. A multi-layer graph neural network is constructed to enable efficient information aggregation and dynamic state inference among nodes. In addition, a perception representation method is designed by fusing local states with global features. This improves each node's ability to perceive the overall system status. The proposed method is evaluated within a customized experimental framework. A dataset featuring heterogeneous task loads and dynamic communication topologies is used. Performance is measured in terms of task completion rate, average latency, load balancing, and transmission efficiency. Experimental results show that the proposed method outperforms mainstream algorithms under various conditions, including limited bandwidth and dynamic structural changes. It demonstrates superior perception capabilities and cooperative scheduling performance. The model achieves rapid convergence and efficient responses to complex system states.

Problem

• Overcoming multi-node perception limitations in distributed systems
• Reducing delayed scheduling response via GNN-based collaboration
• Enhancing dynamic state inference and information aggregation efficiency

Innovation

• GNN-based multi-node collaborative perception mechanism
• Message-passing and state-update modules integration
• Fusion of local states with global features