Serverless Architecture for Machine Learning - Project Report

Aug 2, 2026

Serverless computing is a cloud paradigm that eliminates manual management of infrastructure and services. Built on Function as a Service (FaaS), it orchestrates stateless, event-driven functions for cloud-deployed applications, enabling efficient handling of web services. Machine learning introduces additional demands due to high-compute workloads and large volumes of data, requiring optimized deployment architectures. The work also addresses serverless limitations such as cold starts by evaluating provisioning tools and techniques, deploying an ML model for real-time crisis detection using AWS and GCP, comparing methodologies, and aiming to extend toward a training platform.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages52

Reading Time2 h 11 min

AccessFree to download

Table of Contents

1. Introduction

1.1 Problem Statement

1.2 Motivation

2. Background

3. Related Work

4. Machine Learning

5. Methodologies

6. Results

7. Future Scope

8. Conclusion

References

List of Figures

San Jose State University SJSU ScholarWorks | Master's Projects | Master's Theses and Graduate Research | | --- | --- | | Fall 2023
Serverless Architecture for Machine Learning Ikshaku Goswami
Follow this and additional works at: [https://scholarworks.sjsu.edu/etd_projects](https://scholarworks.sjsu.edu/etd_projects)
Part of the Other Computer Engineering Commons | | Recommended Citation Goswami, Ikshaku, "Serverless Architecture for Machine Learning" (2023) . Master 's Projects. 1336. [https://scholarworks.sjsu.edu/etd_projects/1336](https://scholarworks.sjsu.edu/etd_projects/1336) This Master's Project is brought to you for free and open access by the Master's Theses and Graduate Research at SJSU ScholarWorks. It has been accepted for inclusion in Master's Projects by an authorized administrator of SJSU ScholarWorks. For more information, please contact [scholarworks@sjsu.edu](scholarworks@sjsu.edu). Serverless Architecture for Machine Learning A Project Report Presented to Dr. Robert Chun Department of Computer Science San José State University In Partial Fulfillment Of the Requirements for the Degree Master of Science By Ikshaku Goswami © 2023 Ikshaku Goswami ALL RIGHTS RESERVED 2 The Designated Project Committee Approves the Project Titled Serverless Architecture for Machine Learning by Ikshaku Goswami Approved for the Department of Computer Science San Jose State University December 2023 Dr. Robert Chun Dr. Navrati Saxena Dr. Thomas Austin Department of Computer Science Department of Computer Science Department of Computer Science Acknowledgements I extend my sincere gratitude to Dr. Robert Chun, my esteemed project advisor, for his invaluable guidance, unwavering encouragement, and steadfast support throughout the entire duration of this project. His expertise has been instrumental in shaping the trajectory of my work. I would also like to express my heartfelt thanks to Dr. Navrati Saxena and Dr. Thomas Austin, esteemed members of my committee, for graciously agreeing to be part of this academic journey. Their insightful guidance and constructive feedback have significantly enriched the quality of my research. My deepest appreciation goes to my family; their unwavering support and understanding have been the bedrock of my perseverance. Without them, reaching this milestone would not have been possible. Abstract Serverless computing is an area under cloud computing which does not require individual management of cloud infrastructure and services. It is the groundwork behind Function as a Service or FaaS cloud computing technique. FaaS provides a stateless event-driven orchestration of functions and services for applications deployed in the cloud, without having to manage the servers and other infrastructure resources. This event driven architecture is being well utilized to manage different web-applications and services. Machine learning can bring a unique challenge to serverless computing, as it involves high-intensive tasks which requires voluminous data. In such a scenario it becomes essential to optimize the cloud-deployment architecture to obtain accurate results efficiently. In addition, serverless computing suffers from drawbacks like cold start etc., which further increases the need of researching different serverless provisioning tools and techniques. This research work aims to deploy a machine learning model to detect real-time crisis, using various serverless computing resources provided by notable cloud vendors like Amazon Web Services (AWS) and Google Cloud Platform (GCP) . It also compares among the various methodologies available and later aims to build a training platform for machine learning tasks. Index Terms – Serverless, AWS, Lambda, S3, EFS, EC2, VM, GCF TABLE OF CONTENTS 1. Introduction...........................................................................................................8 1.1 Problem Stateme

FAQs

What problem does serverless architecture aim to solve for cloud applications?

Serverless computing removes the need for individual management of cloud infrastructure and services, leveraging FaaS to orchestrate event-driven functions.

Why is deploying machine learning models challenging in a serverless environment?

Machine learning involves compute-intensive tasks and voluminous data, so deployment architectures must be optimized for accurate and efficient results.

How does the project evaluate serverless approaches for machine learning tasks?

It deploys a machine learning model for real-time crisis detection using AWS and Google Cloud Platform resources, compares available methodologies, and outlines plans to build a training platform.

Aug 2, 2026

Upload by Ava Thompson