Book Image

Distributed Machine Learning with Python

By : Guanhua Wang
Book Image

Distributed Machine Learning with Python

By: Guanhua Wang

Overview of this book

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.
Table of Contents (17 chapters)
Section 1 – Data Parallelism
Section 2 – Model Parallelism
Section 3 – Advanced Parallelism Paradigms

Parameter server architecture

In this section, we will dive into the system architecture of the parameter server paradigm. The domain knowledge requirements for this section are as follows:

  • A Master/Worker architecture in distributed systems
  • Client/Server communication

The parameter server architecture mainly consists of two roles: parameter server and worker. The parameter server can be regarded as the master node in the traditional Master/Worker architecture.

Workers are the computer nodes or GPUs that are responsible for model training. We split the total training data among all the workers. Each worker trains their local model with the training data partition that's been assigned to it.

The duties of parameter server are twofold:

  • Aggregate model updates from all the workers.
  • Update the model parameters held on the parameter server.

The following diagram depicts a simplified parameter server architecture with two workers and one...