Book Image

Intelligent Workloads at the Edge

By : Indraneel Mitra, Ryan Burke
Book Image

Intelligent Workloads at the Edge

By: Indraneel Mitra, Ryan Burke

Overview of this book

The Internet of Things (IoT) has transformed how people think about and interact with the world. The ubiquitous deployment of sensors around us makes it possible to study the world at any level of accuracy and enable data-driven decision-making anywhere. Data analytics and machine learning (ML) powered by elastic cloud computing have accelerated our ability to understand and analyze the huge amount of data generated by IoT. Now, edge computing has brought information technologies closer to the data source to lower latency and reduce costs. This book will teach you how to combine the technologies of edge computing, data analytics, and ML to deliver next-generation cyber-physical outcomes. You’ll begin by discovering how to create software applications that run on edge devices with AWS IoT Greengrass. As you advance, you’ll learn how to process and stream IoT data from the edge to the cloud and use it to train ML models using Amazon SageMaker. The book also shows you how to train these models and run them at the edge for optimized performance, cost savings, and data compliance. By the end of this IoT book, you’ll be able to scope your own IoT workloads, bring the power of ML to the edge, and operate those workloads in a production setting.
Table of Contents (17 chapters)
1
Section 1: Introduction and Prerequisites
3
Section 2: Building Blocks
10
Section 3: Scaling It Up
13
Section 4: Bring It All Together

Exploring the topology of the edge

Solutions built for the edge take on many shapes and sizes. The number of distinct devices included in a solution ranges from one to many. The network layout, compute resources, and budget allowed will drive your architectural and implementation decisions. In an edge machine learning (ML) solution, we should consider the requirements for running ML models. ML models work more accurately when they are custom built for a specific instance of a device, as opposed to one model supporting many physical instances of the same device. This means that as the number of devices supported by an edge ML workload grows, so too will the number of ML models and compute resources required at the edge. There are four topologies to consider when architecting an edge ML solution: star, bus, tree, and hybrid. Here is a description of each of them:

  • Star topology: The Home Base Solutions (HBS) hub device and appliance monitoring kits represent a common pattern...