Review of Observability Driven Security Engineering in Modern Cloud Native Applications

Abstract

One advantage of cloud-native apps is that they form the backbone of decentralised computer networks. The characteristics of cloud-native apps are scalability, flexibility, and resilience. However, the dynamics of microservices, containers, and Kubernetes architectures create many difficulties in system monitoring and identification of its vulnerabilities. The traditional approach to system monitoring based on the use of static pre-defined rules and threshold alerts does not provide enough information and is often unable to identify anomalies or potential dangers. The aim of this paper is to present an enhanced AI-enabled cloud-native observability framework for improving system monitoring, detection of system vulnerabilities, and decision-making processes. In addition, the existing technologies and techniques for cloud-native observability, intelligent monitoring, and security such as distributed tracing, eBPF, open Telemetry, artificial intelligence, and machine learning analyzed. Comparative review of the existing technologies demonstrates their advantages, disadvantages, and application possibilities in cloud-native architecture. It shows how the intelligent approach to monitoring can improve the fault detection and optimization of resource consumption, as well as enhance security monitoring and autonomous operation of cloud. Finally, the problems faced in the current state of research discussed and future research directions outlined.