Abstract

Cloud-native architectures and serverless computing have revolutionized application deployment, improving scalability, reducing costs, and increasing operational flexibility. Optimizing performance, resource allocation, and automating workflows continues to be a difficult challenge. This paper discusses the latest developments in serverless computing, focusing on its importance for building scalable applications. Key developments in container-based platforms, workflow execution, and function choreography are explored. Furthermore, we evaluate the combination of serverless computing with AI, scientific workflows, and high performance computing. The research study uncovers performance limitations including cold start latency and shortcomings in resource allocation, whilst also examining methods for streamlining workflows and redistributing workload. Additionally, it considers issues such as cold start latency and resource provisioning inefficiencies. This study synthesizes existing research to offer insights into the advantages, obstacles, and potential future paths for improving cloud-native and serverless infrastructure designs.