Evaluating Serverless Java Architectures: Performance Insights for Full-Stack AI-Driven Cloud Solutions

EasyChair Preprint 15016

Abstract

In the rapidly evolving landscape of cloud computing, serverless architectures have emerged as a compelling model for deploying applications, particularly in the context of full-stack AI-driven solutions. This article presents a comprehensive evaluation of serverless Java architectures, focusing on performance metrics critical to modern cloud-based applications. By analyzing various serverless frameworks and their implementations, we offer insights into how these architectures handle the computational demands of AI-driven processes. The study explores key performance indicators such as latency, scalability, and resource utilization, providing a comparative analysis between serverless Java and traditional server-based approaches. Through empirical testing and benchmarking across diverse scenarios, our findings highlight the strengths and limitations of serverless environments in supporting complex, full-stack AI applications. This evaluation aims to guide developers and architects in optimizing their serverless deployments to achieve robust performance and efficiency in the cloud.

Keyphrases: Cloud, Performance, architects, developers, efficiency, evaluation, robust

@booklet{EasyChair:15016,
  author    = {Adeyeye Barnabas},
  title     = {Evaluating Serverless Java Architectures: Performance Insights for Full-Stack AI-Driven Cloud Solutions},
  howpublished = {EasyChair Preprint 15016},
  year      = {EasyChair, 2024}}