Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Jaime Sebastian Burbano, Arnova Abdullah, Eldiyar Zhantileuov, Mohan Liyanage, Rolf Schuster

TL;DR

This paper tackles reliable, low-latency edge server selection under time-varying network conditions for latency-sensitive embedded applications. It introduces MO-HAN, a lightweight, prediction-guided server selector that fuses end-to-end latency forecasts from a rational delay model with an online reliability score and hysteresis-based switching to avoid oscillations. The approach decomposes into Prediction-Guided Server Selection, Adaptive Reliability Modeling, and Switching-Aware Handover, with empirical validation on a custom testbed showing reduced mean and tail latency and roughly 50% fewer handovers than opportunistic strategies. The method is practical for resource-constrained devices due to its reliance on passive measurements and absence of heavy learning infrastructure, and it can be extended to multi-objective optimization and federated edge scenarios.

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.

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

TL;DR

This paper tackles reliable, low-latency edge server selection under time-varying network conditions for latency-sensitive embedded applications. It introduces MO-HAN, a lightweight, prediction-guided server selector that fuses end-to-end latency forecasts from a rational delay model with an online reliability score and hysteresis-based switching to avoid oscillations. The approach decomposes into Prediction-Guided Server Selection, Adaptive Reliability Modeling, and Switching-Aware Handover, with empirical validation on a custom testbed showing reduced mean and tail latency and roughly 50% fewer handovers than opportunistic strategies. The method is practical for resource-constrained devices due to its reliance on passive measurements and absence of heavy learning infrastructure, and it can be extended to multi-objective optimization and federated edge scenarios.

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.

Paper Structure

This paper contains 13 sections, 7 equations, 4 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related work
  3. System Model
  4. Network Monitoring and Latency Prediction
  5. Adaptive Reliability and Moderated Handover Modeling
  6. Experimental Setup and Results
  7. Testbed Overview
  8. Components and Functionality
  9. Data Collection
  10. Operational Flow
  11. Server Selection Algorithms
  12. Results
  13. Conclusions

Figures (4)

  • Figure 1: Block diagram of the proposed server selection methodology. In time-varying conditions, sporadic congestion in network paths to a certain edge server leads to elevated latency.
  • Figure 2: Experimental setup with dynamic load and real-time monitoring.
  • Figure 3: Server selection by implementing the proposed MO-HAN algorithm.
  • Figure 4: Cumulative distributed function of end-to-end latency with the different server selection algorithms.