Date of Graduation

Summer 8-9-2024

Document Type

Thesis

Degree Name

Master of Computer and Information Science (MCIS)

Department

Computational, Engineering and Mathematical Science

Thesis Chair

Dr. Jeong Yang

Abstract

With the enhanced computing capabilities and accessibility to cloud resources, major cloud computing providers such as Google Cloud Platform (GCP), Amazon Web Services (AWS), and Microsoft Azure offer Machine Learning (ML) and AI services. Their primary purpose is to provide efficiency, scalability, and adaptability in modern software development and IT operations while reducing overall costs and operational complexity. However, prospective customers of the services often question which ML-AI service will best suit their organizational and business needs. This study compares and analyzes the usability, performance, and cost-efficiency of deploying Machine Learning (ML) models across three cloud platforms: GCP, AWS, and Azure. Specifically, it examines BigQuery ML and Vertex AI in Google Cloud, Redshift and SageMaker in AWS, and Azure Machine Learning. Using MIMIC-IV datasets of hospitalized patients, regression models were deployed on each platform to predict mortality and disease progression. The analysis results revealed that Google’s BigQuery ML offers good usability with excellent documentation and a moderate learning curve, making it suitable for SQL-savvy users and large-scale data analytics tasks. Google’s Vertex AI, while more flexible, incurs relatively higher costs. AWS SageMaker’s costs are unpredictable due to its parallel training processes, while Azure ML, though requiring complex authorization settings, is accessible for novices. These findings provide valuable insights for selecting an appropriate AI-ML service for deploying similar ML models, providing the strengths and limitations of each platform, aiding in informed decision-making for cloud-based ML deployments.

Download

Share

COinS