Dynamic Edge Server Selection in Time-Varying Environments: A Reliability-Aware Predictive Approach

2025-11-13

AI Summary

To address high end-to-end latency and frequent server handovers caused by edge server selection under dynamic network congestion, this paper proposes a lightweight predictive selection method. The method integrates passive measurements—including arrival rate, utilization, and load—with an exponentially modulated rational delay model to achieve low-overhead latency prediction. It further introduces reliability-aware dynamic scoring and a hysteresis-based adaptive handover mechanism to suppress unnecessary switches while maintaining low latency. Compared to baseline approaches, the proposed method reduces both average and tail latency significantly and cuts handover frequency by nearly 50%, without requiring intrusive monitoring or complex machine learning models. Designed for resource-constrained embedded edge devices, it jointly optimizes responsiveness, stability, and deployment efficiency.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Planning, Routing, and Scheduling: Model-Based Reasoning
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Security and Privacy: Large-scale security measurements

Abstract

Latency-sensitive embedded applications increasingly rely on edge computing, yet dynamic network congestion in multi-server architectures challenges proper edge server selection. This paper proposes a lightweight server-selection method for edge applications that fuses latency prediction with adaptive reliability and hysteresis-based handover. Using passive measurements (arrival rate, utilization, payload size) and an exponentially modulated rational delay model, the proposed Moderate Handover (MO-HAN) method computes a score that balances predicted latency and reliability to ensure handovers occur only when the expected gain is meaningful and maintain reduced end-to-end latency. Results show that MO-HAN consistently outperforms static and fair-distribution baselines by lowering mean and tail latencies, while reducing handovers by nearly 50% compared to pure opportunistic selection. These gains arise without intrusive instrumentation or heavy learning infrastructure, making MO-HAN practical for resource-constrained embedded devices.

Problem

• Selecting optimal edge servers under dynamic network congestion
• Balancing latency prediction with reliability for server handovers
• Reducing unnecessary handovers while maintaining low end-to-end latency

Innovation

• Lightweight server selection fuses latency prediction
• Balances predicted latency and reliability with scores
• Uses passive measurements without heavy learning infrastructure