Publication Date

Spring 2021

Degree Type

Master's Project

Degree Name

Master of Science in Computer Science (MSCS)

Department

Computer Science

First Advisor

Robert Chun

Second Advisor

Fabio Di Troia

Third Advisor

Soon Tee Teoh

Keywords

Serverless Computing, Distributed Machine Learning

Abstract

Machine learning has been trending in the domain of computer science for quite some time. Newer and newer models and techniques are being developed every day. The adoption of cloud computing has only expedited the process of training machine learning. With its variety of services, cloud computing provides many options for training machine learning models. Leveraging these services is up to the user. Serverless computing is an important service offered by cloud service providers. It is useful for short tasks that are event-driven or periodic. Machine learning training can be divided into short tasks or batches to take advantage of this. Due to the nature of serverless computing, there are certain limitations imposed by the cloud service provider such as execution time and memory. This research proposes standalone solutions to overcome the challenges faced by serverless computing in training machine learning models. The research further combines these individual solutions and proposes a system for leveraging serverless computing for training a machine learning model that incorporates distributed machine learning.

Download

Share

COinS