Article: Dynamic load-balancing optimisation with bidirectional edge detection under multi-scale feature fusion Journal: International Journal of Data Mining and Bioinformatics (IJDMB) 2026 Vol.30 No.6 pp.49 - 70 Abstract: This paper proposes a scheduling optimisation framework GCN-EDQNet that integrates multi-scale feature fusion, bidirectional edge detection, and graph convolutional network (GCN). Data centre resources are modelled as a graph over distributed nodes; a multi-scale module builds hierarchical representations to capture spatial heterogeneity in resource distributions. A bidirectional edge-detection subnetwork then identifies scheduling-sensitive regions and produces an edge heatmap - assigning higher weights to edges connecting nodes with high load variation - that guides the GCN to prioritise structurally salient, mutation-prone areas. This explicit weighting mechanism enables the GCN to focus on bottleneck-prone regions and improve structure-aware feature learning. Finally, a reinforcement learning strategy enables adaptive task allocation and migration. Experiments on two public datasets show that GCN-EDQNet outperforms conventional approaches in task completion time, load variance, scheduling success rate, and energy efficiency. These results highlight a structure-aware, intelligent paradigm for data centre resource scheduling with clear theoretical and practical value. Inderscience Publishers - linking academia, business and industry through research

Title: Dynamic load-balancing optimisation with bidirectional edge detection under multi-scale feature fusion

Authors: Xi Song; Bo Chen; Lin Tian; Xiaodong Wei; Kai Song

Addresses: SGIT-UNI (Lanzhou) Cloud Data Technology Co., Ltd., No. 628, Xijin East Road, Qilihe District, 730050, Lanzhou, Gansu Province, China ' SGIT-UNI (Lanzhou) Cloud Data Technology Co., Ltd., No. 628, Xijin East Road, Qilihe District, 730050, Lanzhou, Gansu Province, China ' SGIT-UNI (Lanzhou) Cloud Data Technology Co., Ltd., No. 628, Xijin East Road, Qilihe District, 730050, Lanzhou, Gansu Province, China ' SGIT-UNI (Lanzhou) Cloud Data Technology Co., Ltd., No. 628, Xijin East Road, Qilihe District, 730050, Lanzhou, Gansu Province, China ' SGIT-UNI (Lanzhou) Cloud Data Technology Co., Ltd., No. 628, Xijin East Road, Qilihe District, 730050, Lanzhou, Gansu Province, China

Abstract: This paper proposes a scheduling optimisation framework GCN-EDQNet that integrates multi-scale feature fusion, bidirectional edge detection, and graph convolutional network (GCN). Data centre resources are modelled as a graph over distributed nodes; a multi-scale module builds hierarchical representations to capture spatial heterogeneity in resource distributions. A bidirectional edge-detection subnetwork then identifies scheduling-sensitive regions and produces an edge heatmap - assigning higher weights to edges connecting nodes with high load variation - that guides the GCN to prioritise structurally salient, mutation-prone areas. This explicit weighting mechanism enables the GCN to focus on bottleneck-prone regions and improve structure-aware feature learning. Finally, a reinforcement learning strategy enables adaptive task allocation and migration. Experiments on two public datasets show that GCN-EDQNet outperforms conventional approaches in task completion time, load variance, scheduling success rate, and energy efficiency. These results highlight a structure-aware, intelligent paradigm for data centre resource scheduling with clear theoretical and practical value.

Keywords: dynamic load balancing; graph neural networks; GNNs; edge detection; data centre scheduling optimisation.

DOI: 10.1504/IJDMB.2026.153899

International Journal of Data Mining and Bioinformatics, 2026 Vol.30 No.6, pp.49 - 70

Received: 31 Oct 2025
Accepted: 11 Feb 2026
Published online: 29 May 2026 *

Title: Dynamic load-balancing optimisation with bidirectional edge detection under multi-scale feature fusion

Keep up-to-date