QoS-Aware Load Balancing in the Computing Continuum via Multi-Player Bandits

TL;DR

This paper introduces QEdgeProxy, a decentralized load balancer for the Computing Continuum that uses multi-player bandits and KDE to meet per-client QoS requirements effectively in dynamic, latency-sensitive environments.

Contribution

It presents a novel QoS-aware load balancing algorithm based on Multi-Player Multi-Armed Bandits with adaptive exploration, tailored for the Computing Continuum environment.

Findings

QEdgeProxy outperforms baseline methods in QoS satisfaction.

It adapts effectively to load surges and environment changes.

The implementation demonstrates practical viability on Kubernetes clusters.

Abstract

As computation shifts from the cloud to the edge to reduce processing latency and network traffic, the resulting Computing Continuum (CC) creates a dynamic environment where meeting strict Quality of Service (QoS) requirements and avoiding service instance overload becomes challenging. Existing methods often prioritize global metrics and overlook per-client QoS, which is crucial for latency-sensitive and reliability-critical applications. We propose QEdgeProxy, a decentralized QoS-aware load balancer that acts as a proxy between IoT devices and service instances in the CC. We formulate the load balancing problem as a Multi-Player Multi-Armed Bandit (MP-MAB) with heterogeneous rewards: Each load balancer autonomously selects service instances to maximize the probability of meeting its clients' QoS requirements by using Kernel Density Estimation (KDE) to estimate QoS success probabilities. Our load-balancing algorithm also incorporates an adaptive exploration mechanism to recover rapidly from performance shifts and non-stationary conditions. We present a Kubernetes-native QEdgeProxy implementation and evaluate it on an emulated CC testbed deployed on a K3s cluster with realistic network conditions and a latency-sensitive edge-AI workload. Results show that QEdgeProxy significantly outperforms proximity-based and reinforcement-learning baselines in per-client QoS satisfaction, while adapting effectively to load surges and changes in instance availability.