Analyzing the Performance and Cost-Efficiency of Deploying Machine Learning Models in Cloud

Hongyu Wang

Document Type

Conference Proceeding

Publication Date

2024

Abstract

This study presents the results and findings on the analysis of the performance and cost-efficiency to deploy different Machine Learning (ML) models in Cloud. It specifically examines the operational dynamics and cost-efficiency of using machine learning functions within the integrated database vs external machine learning services for model training and prediction. Focusing on the Google Cloud Platform (GCP), the research employs a comparative approach to assess how these deployment strategies influence the scalability, resource utilization, and overall cost implications in cloud environments. The anticipated outcomes of this investigation aim to provide insightful guidance about the performance and cost efficiency of various cloud-based ML deployment methodologies. The findings are expected to offer significant contributions to the field, guiding the optimization of ML model deployments in cloud settings, and aiding stakeholders in making informed decisions for leveraging cloud technologies in ML, and further AI (Artificial Intelligence) applications.

Comments

Learning and Living
BLH 362

Link to Full Text

Share

COinS