DMSA: A Decentralized Microservice Architecture for Edge Networks

TL;DR

This paper addresses the fragility of centralized microservice architectures in edge networks by introducing DMSA, a decentralized MSA that pushes scheduling to edge-node agents and redesigns discovery, monitoring, and scheduling modules for precise deployment awareness and low overhead. It introduces PSMS for status synchronization, CAPA and EMA-based bandwidth estimation, and a dynamic weighted multilevel load balancing (DWMLB) that prioritizes reliability, priority, and response delay using multi-port, zero-copy forwarding. The authors provide a detailed implementation on a Raspberry Pi/Orange Pi testbed, compare against Nautilus, Least Connects, and Round Robin, and show substantial improvements in latency, execution success rates, and resilience to link failures and network fluctuations. The combination of decentralized scheduling, lightweight monitoring, and efficient forwarding demonstrates significant practical impact for real-time edge applications and dynamic microservice dependencies. The work also lays out a concrete platform for future exploration of adaptive deployment strategies at the edge.

Abstract

The dispersed node locations and complex topologies of edge networks, combined with intricate dynamic microservice dependencies, render traditional centralized microservice architectures (MSAs) unsuitable. In this paper, we propose a decentralized microservice architecture (DMSA), which delegates scheduling functions from the control plane to edge nodes. DMSA redesigns and implements three core modules of microservice discovery, monitoring, and scheduling for edge networks to achieve precise awareness of instance deployments, low monitoring overhead and measurement errors, and accurate dynamic scheduling, respectively. Particularly, DMSA has customized a microservice scheduling scheme that leverages multi-port listening and zero-copy forwarding to guarantee high data forwarding efficiency. Moreover, a dynamic weighted multi-level load balancing algorithm is proposed to adjust scheduling dynamically with consideration of reliability, priority, and response delay. Finally, we have implemented a physical verification platform for DMSA. Extensive empirical results demonstrate that compared to state-of-the-art and traditional scheduling schemes, DMSA effectively counteracts link failures and network fluctuations, improving the service response delay and execution success rate by approximately $60\% \sim 75\%$ and $10\%\sim15\%$, respectively.

Figures (17)

• Figure 1: Traditional centralized microservice architecture.
• Figure 2: The performance impact of microservice discovery, monitoring, and scheduling. (a) Microservice Invocations. (b) Two different schemes for the depicted microservice invocation.
• Figure 3: The decentralized microservice architecture.
• Figure 4: The request forwarding process of DMSA.
• Figure 5: The relationships among DMSA's key functional modules.