The Impact of using MLOps and DevOps on Container based Applications: A Survey

Document Type : Original Article

Authors

1 Helwan

2 Department of Information Systems, Faculty of Computers and Artificial Intelligence, Helwan University, Egypt

3 Information System, Faculty of Computer and Artificial Intelligence, Helwan University, Cairo, Egypt

Abstract

Developing and implementing the machine learning applications as quickly as feasible is the aim of commercial machine learning (ML) initiatives. A lot of machine learning trials, however, fall short of their requirements and expectations because automating and operationalizing the machine learning systems is so challenging. MLOps, a model for machine learning operations, deals with this. There are several components to MLOps, including development culture, collections of concepts, and best practices. The consequences of MLOps for academics and professionals are, however, yet unknown. As a result, this study provides a detailed description of the underlying concepts, elements, as well architecture, to aid in the development of usable software based on ML methods that can make MLOps models simple to monitor, integrate, and scale securely to help save maintenance costs and improve the number of software deployments. The study draws attention to the remaining problems in the area before it is done.

Keywords

Main Subjects

Files

Share

How to cite

Statistics