Risk-Aware and Stable Edge Server Selection Under Network Latency SLOs

2026-04-23

AI Summary

This work addresses the challenge of dynamic server selection in edge computing under stringent latency SLOs, where decisions must jointly manage tail-risk violations and switching stability. The authors propose a lightweight, interpretable decision framework that uniquely co-optimizes tail-risk control and switching stability: it estimates SLO violation risk using normal approximation and the Cantelli inequality, while incorporating a hysteresis mechanism to suppress excessive switching. Experimental results under a 0.5-second SLO demonstrate that, compared to a baseline relying solely on mean latency, the proposed method reduces deadline miss rate from 39% to 34%, cuts switching frequency by 88% (down to 5.5%), and maintains average latency stably at 0.45 seconds, thereby significantly enhancing both system robustness and efficiency.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Reasoning under Uncertainty: Stochastic Optimization
• Planning, Routing, and Scheduling: Scheduling under Uncertainty

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Cloud, edge and content delivery systems for the Web

Abstract

We present a lightweight and interpretable decision framework for dynamic edge server selection in latency-critical applications that explicitly accounts for tail risk and switching stability. Each candidate server is characterised by predictive mean and uncertainty summaries of network latency, which are used to estimate the risk of service-level objective (SLO) violations and to guide selection. Risk is evaluated using a tight Normal approximation complemented by a conservative Cantelli bound, while percentile-based scoring coupled with hysteresis stabilizes decisions and suppresses oscillatory switching under short-lived network fluctuations. Experimental results on a multi-server edge testbed with a strict SLO of $τ= 0.5$\,s show that the proposed approach reduces the deadline-miss rate from 39\% to 34\% compared to a mean-only baseline, while reducing switching frequency from 46\% to 5.5\% ($\approx$88\% reduction) and maintaining sub-SLO average latency ($\approx$0.45\,s). These results demonstrate that explicit risk evaluation combined with stability-preserving control enables practical and robust adaptive server selection in dynamic edge environments.

Problem

• edge server selection
• network latency
• service-level objective (SLO)
• risk-awareness
• switching stability

Innovation

• risk-aware server selection
• latency SLO
• tail risk
• hysteresis control
• edge computing