MLOps with Red Hat OpenShift: A cloud-native approach to machine learning operations

A practical approach to implementing MLOps with Red Hat OpenShift. See how Red Hat and its technology partners such as Pachyderm and Intel OpenVINO provide a consistent platform across clouds and on-premises hardware.

Key Features

• Understand the concepts of MLOps and machine learning project lifecycle
• Experience provisioning an MLOps ecosystem using a step-by-step guide
• Build a complete ML workflow from using OpenShift Data Science

Book Description

MLOps with OpenShift provides a practical approach to implementing MLOps workflows on the OpenShift platform. The book begins by introducing key MLOps concepts, including data preparation, model training, and deployment. It then provides an overview of OpenShift, covering the basic blocks of OpenShift, such as containers, pods and operators.

With the basics covered, the book then dives into MLOps workflows on the OpenShift platform. Readers will learn how popular machine learning frameworks are used to train and test models on the platform.

The book will cover Red Hat OpenShift Data Science, an open-source data science and machine learning platform designed to run on the OpenShift platform. Red Hat OpenShift Data Science and partner components provide the building blocks to build and manage data pipelines and deploy and monitor machine learning models. Pachyderm and Intel OpenVINO are two such partner components covered in the book.

By the end of the book, readers will have a solid understanding of MLOps concepts and best practices, as well as the skills and knowledge needed to implement MLOps workflows on the OpenShift platform.

What you will learn

• Gain a deep understanding of the key concepts and best practices of MLOps
• Learn different components that cover the essential MLOps workflows for developing, training, testing, deploying, and monitoring machine learning models at scale
• Implement complete MLOps workflows on the Red Hat OpenShift platform
• Configure different components of an MLOps platform on an OpenShift cluster
• Learn how different teams can collaborate over the OpenShift platform
• Learn how to operate data science and machine learning workloads in OpenShift

Who This Book Is For

MLOps and DevOps engineers, Data architects and scientists, and developers who want to learn MLOps and its components are the primary audiences. They will learn how to integrate these components for building a complete machine learning lifecycle on the Red Hat OpenShift Data Science platform. Whether you are a data scientist, machine learning engineer, or software developer, "MLOps with OpenShift" is the essential resource for building scalable and efficient machine learning workflows on the OpenShift container platform.

Table of Contents

1. Introduction to MLOps and OpenShift
2. The Machine Learning Lifecycle
3. Provisioning an MLOps platform in the Cloud
4. Building Machine Learning Models
5. Embedding ML Models into the Applications
6. Deploying ML Models as a Service
7. Operating ML workloads
8. Building MLOps platform for success

Ross Brigoli is a consulting architect at Red Hat, where he focuses on designing and delivering solutions around microservices architecture, DevOps, and MLOps with Red Hat OpenShift for various industries. He has two decades of experience in software development and architecture.

Faisal Masood is a cloud transformation architect at AWS. Faisal's focus is to assist customers in refining and executing strategic business goals. Faisal main interests are evolutionary architectures, software development, ML lifecycle, CD and IaC. Faisal has over two decades of experience in software architecture and development.