Reducing time cost in machine learning leads to a shorter waiting time for model training and a faster model updating cycle. Distributed machine learning enables machine learning practitioners to shorten model training and inference time by orders of magnitude. With the help of this practical guide, you'll be able to put your Python development knowledge to work to get up and running with the implementation of distributed machine learning, including multi-node machine learning systems, in no time. You'll begin by exploring how distributed systems work in the machine learning area and how distributed machine learning is applied to state-of-the-art deep learning models. As you advance, you'll see how to use distributed systems to enhance machine learning model training and serving speed. You'll also get to grips with applying data parallel and model parallel approaches before optimizing the in-parallel model training and serving pipeline in local clusters or cloud environments. By the end of this book, you'll have gained the knowledge and skills needed to build and deploy an efficient data processing pipeline for machine learning model training and inference in a distributed manner.
What you will learn
Deploy distributed model training and serving pipelines
Get to grips with the advanced features in TensorFlow and PyTorch
Mitigate system bottlenecks during in-parallel model training and serving
Discover the latest techniques on top of classical parallelism paradigm
Explore advanced features in Megatron-LM and Mesh-TensorFlow
Use state-of-the-art hardware such as NVLink, NVSwitch, and GPUs
Who this book is for
This book is for data scientists, machine learning engineers, and ML practitioners in both academia and industry. A fundamental understanding of machine learning concepts and working knowledge of Python programming is assumed. Prior experience implementing ML/DL models with TensorFlow or PyTorch will be beneficial. You'll find this book useful if you are interested in using distributed systems to boost machine learning model training and serving speed.
Table of Contents
Splitting Input Data
Parameter Server and All-Reduce
Building a Data Parallel Training and Serving Pipeline
Bottlenecks and Solutions
Splitting the Model
Pipeline Input and Layer Split
Implementing Model Parallel Training and Serving Workflows
Achieving Higher Throughput and Lower Latency
A Hybrid of Data and Model Parallelism
Federated Learning and Edge Devices
Elastic Model Training and Serving
Advanced Techniques for Further Speed-Ups