Operational AI with Docker : Deploy, Scale, and Operate Agentic AI Services with Docker and Kubernetes.

Format:

Book

Author/Creator:

Raina, Ajeet Singh.

Contributor:

Manvar, Harsh.

Language:

English

Subjects (All):

Database management.

Application software--Development--Computer programs.

Application software.

Physical Description:

1 online resource (390 p.)

Place of Publication:

Birmingham : Packt Publishing, Limited, 2026.

Summary:

Run production-grade GenAI workloads by containerizing, serving, and scaling LLMs, agents, and multi-model pipelines with Docker, MCP, and Kubernetes for cloud platforms Key Features Deploy and operate local and edge-friendly LLM inference using Docker Model Runner and an OpenAI-compatible API Orchestrate multi-model and multi-agent workloads...

Contents:

Cover

Operational AI with Docker

Deploy, scale, and operate agentic AI services with Docker and Kubernetes

Contributors

About the authors

About the reviewers

Table of Contents

Preface

Who this book is for

What this book covers

To get the most out of this book

Download the example code files

Download the color images

Conventions used

Get in touch

Free benefits with your book

How to Unlock

Share your thoughts

Part 1

Foundations of AI Containerization

1

Docker Desktop

The Runtime Foundation for AI/ML Workflows

Technical requirements

What is Docker and why does it matter for AI/ML?

Containers vs VMs

How does Docker architecture work: Images, containers, and registries?

Images: Immutable, layered build artifacts

Tags vs digests

Containers: Runtime instances and lifecycle

Registries: Distribution and collaboration

How do you set up Docker Desktop for local development?

Installing Docker Desktop

Verifying the installation

Why settings matter for AI/ML workflows

Practical settings for AI workloads

Resource planning: CPU, memory, and local constraints

Key settings to configure early

Run a simple smoke test container

How to run your first container?

Step 1: Run your first container (hello-world)

Step 2: Run a web container with port mapping

Step 3: Inspect and interact via Docker Desktop

Step 4: Clean up containers and images

How do the basic Docker commands work: run, build, ps, images, and logs?

Step 1: Start containers with docker run

Step 2: Check what is running with docker ps

Step 3: See available images with docker images

Step 4: Debug with docker logs

Step 5: Build your own image with docker build

Step 6: Keep builds fast with .dockerignore

Step 7: Use layer caching effectively

Step 8: Debug build failures

Step 9: Understand build output

Step 10: Apply best practices for ML Dockerfiles

What is Docker's role in modern AI workflows?

Two recurring threads in this book

What actually makes a workflow reproducible?

Hands-on: Containerize a tiny training run

Step 1: Set up the project structure

Step 2: Build and run the container

Making outputs usable: Mounting artifacts

Extending the same idea to a real workflow

Running ML stacks without installing them locally

GPU-enabled version (using host GPU)

Notebook workflow: Run Jupyter inside a container

From local workflow to team workflow

Working with registries

A minimal registry workflow (push/pull)

Operational considerations: image size and structure

Image size awareness

Inspecting image layers

Edge deployment thread: smaller images, fewer assumptions

From jobs to services: A quick mental model

Common pitfalls (and how to avoid them)

Reproducibility checklist

Summary

Notes:

Description based upon print version of record.

Get this book's PDF version and more

OCLC-licensed vendor bibliographic record.

ISBN:

9781807301095

OCLC:

1587902534