Microservices (MSs)-based architectures have become the de facto standard for designing and implementing edge computing applications. In particular, by leveraging Network Performance Metrics (NPMs) coming from the Radio Access Network (RAN) and sharing context-related information, AI-driven MSs have demonstrated to be highly effective in optimizing RAN performance. In this context, this work addresses the critical challenge of ensuring efficient data sharing and consistency by proposing a holistic platform that regulates the collection and usage of NPMs. We first introduce two reference platform architectures and detail their implementation using popular, off-the-shelf database solutions. Then, to evaluate and compare such architectures and their implementation, we develop PACE, a highly configurable, scalable, MS-based emulation framework of producers and consumers of NPMs, capable of realistically reproducing a broad range of interaction patterns and load dynamics. Using PACE on our cloud computing testbed, we conduct a thorough characterization of various NPM platform architectures and implementations under a spectrum of realistic edge traffic scenarios, from loosely coupled control loops to latency- and mission- critical use cases. Our results reveal fundamental trade-offs in stability, availability, scalability, resource usage, and energy footprint, demonstrating how PACE effectively enables the identification of suitable platform solutions depending on the reference edge scenario and the required levels of reliability and data consistency.

Stateful migration has emerged as the dominant technology to support microservice mobility at the network edge while ensuring a satisfying experience to mobile end users. This work addresses two pivotal challenges, namely, the implementation and the orchestration of the migration process. We first introduce a novel framework that efficiently implements stateful migration and effectively orchestrates the migration process by fulfilling both network and application KPI targets. Through experimental validation using realistic microservices, we then show that our solution (i) greatly improves migration performance, yielding up to 77% decrease of the migration downtime with respect to the state of the art, and (ii) successfully addresses the strict user QoE requirements of critical scenarios featuring latency-sensitive microservices. Further, we consider two practical use cases, featuring, respectively, a UAV autopilot microservice and a multi-object tracking task, and demonstrate how our framework outperforms current state-of-the-art approaches in configuring the migration process and in meeting KPI targets.

Stateful migration has emerged as the dominant technology to support microservice mobility at the network edge while meeting the end users' QoE requirements. In this context, our work addresses the two pivotal challenges of implementing and orchestrating the migration process. We first introduce a novel orchestration framework that efficiently realizes stateful migration and effectively orchestrates the migration process by fulfilling both network and application KPI targets. Then, through experimental validation using realistic microservices, we show that our solution improves migration performance, yielding up to 80 % decrease of the migration downtime with respect to the state of the art. Finally, we demonstrate that our framework can be exploited to successfully address critical scenarios featuring latency-sensitive microservices and strict user QoE requirements.

Stateful migration has emerged as the key solution to support latency-sensitive microservices at the edge while ensuring a satisfying experience for mobile users. In this paper, we address two relevant issues affecting stateful migration, namely, the migration of containerized microservices and that of the associated data connection. We do so by first introducing a novel network solution, based on OvS, that permits to preserve the established connection with mobile end users upon migrating a microservice. Then, using Podman and CRIU, we experimentally characterize the fundamental migration KPIs, i.e., migration duration and microservice downtime, and we devise an analytical model that, accounting for all the relevant real-world aspects of stateful migration, provides an accurate upper bound on such KPIs. We validate our model using real-world microservices, namely, MQTT Broker and Memcached, and show that it can predict KPIs values with an error that is up to 99.7% smaller than that yielded by the state of the art. Finally, we consider a UAV controller as relevant microservice use case and demonstrate how our model can be exploited to effectively configure the system parameters so that the required QoE level is met.

To support latency sensitive microservices at the edge, stateful container migration has gathered momentum as a key solution to ensure a satisfying experience to mobile users. In this paper, we first investigate experimentally the stateful migration process, by using state-of-the-art tools, namely, Podman and CRIU. We then characterize the main migration KPIs, i.e., migration duration and downtime, and develop an analytical model that can effectively assess whether stateful migration is feasible while meeting the user's QoE requirements. Importantly, our model is validated using real-world microservices and, by accounting for all relevant real-world aspects of stateful migration, significantly outperforms state-of-the-art models.