Scalable Edge-to-Cloud Architecture for Enterprise Applications Using Microservices and Cloud-Native Platforms

Abstract

The rapid growth of enterprise applications is a major factor in the creation of architectures that can operate fluidly both at the edge and in the cloud. Scalable cloud-native platforms and microservices have generally been a great support in extending the flexibility and the application side of the users, but the appearance of latency-sensitive workloads and the rise in data volumes have unwrapped the drawbacks of traditional cloud-centric architectural models. Enterprises are facing challenges such as how to scale under volatile workloads, dealing with high latencies on the network when performing real-time processing, and orchestrating distributed services in their heterogeneous edge and cloud infrastructures which have become complex due to the differences in the infrastructures. An architectural approach that still gives centralized control and scalability in the cloud while at the same time balancing localized processing at the edge is needed to eliminate such problems. This paper proposes a microservices-based and cloud-native principle-based scalable edge-to-cloud architecture for enterprise applications. The proposed system fragments enterprise workloads into microservices that have a loose connection, and, therefore, they can be dynamically deployed across edge nodes and cloud platforms, thus enabling low-latency processing, efficient resource utilization, and centralized orchestration without any restriction due to the locality. The idea involves devising a multi-layered architectural model that represents the containerization, service orchestration, and communication mechanisms at the edge and cloud tiers and, subsequently, validation through a representative enterprise case study. Various performance metrics such as response time, scalability, fault tolerance, and system reliability are checked under different workload variations. The charts signify the scenarios as a perfect fit for the setup which involves a great amount of work divided among different nodes be it cloud or edge nodes thus achieving scalability and cutting down the data processing time. The architecture proposed at the edge-to-cloud level facilitates the seamless transition of services and the efficient execution of workloads without negatively affecting manageability or security. The main point of this paper is to introduce an actual, enterprise-ready edge-to-cloud model that acts as a link between theoretical cloud native concepts and real-world deployment requirements.