D2Q Synchronizer: Distributed SDN Synchronization for Time Sensitive Applications

2025-08-15

AI Summary

To address the lack of joint optimization of network cost and user latency in existing controller synchronization strategies for distributed SDN, this paper proposes a reinforcement learning–driven multi-domain state synchronization and task scheduling framework. The method jointly models network performance metrics—including bandwidth utilization and computational overhead—with user QoS requirements, particularly end-to-end latency for delay-sensitive applications, enabling dynamic cross-domain task offloading and global state synchronization. Operating within an integrated distributed SDN and edge computing architecture, the approach balances long-term operational cost minimization with stringent real-time guarantees. Experimental evaluation demonstrates that, compared to heuristic and state-of-the-art learning-based baselines, the proposed strategy reduces network cost by 45% and 10%, respectively, while satisfying end-to-end latency constraints for 100% of tasks.

Technology Category

• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Planning/Scheduling and Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

In distributed Software-Defined Networking (SDN), distributed SDN controllers require synchronization to maintain a global network state. Despite the availability of synchronization policies for distributed SDN architectures, most policies do not consider joint optimization of network and user performance. In this work, we propose a reinforcement learning-based algorithm called D2Q Synchronizer, to minimize long-term network costs by strategically offloading time-sensitive tasks to cost-effective edge servers while satisfying the latency requirements for all tasks. Evaluation results demonstrate the superiority of our synchronizer compared to heuristic and other learning policies in literature, by reducing network costs by at least 45% and 10%, respectively, while ensuring the QoS requirements for all user tasks across dynamic and multi-domain SDN networks.

Problem

• Optimize distributed SDN synchronization for time-sensitive applications
• Minimize network costs while meeting task latency requirements
• Improve performance via reinforcement learning-based task offloading

Innovation

• Reinforcement learning-based synchronization algorithm
• Strategic offloading to cost-effective edge servers
• Minimizes long-term network costs dynamically