The Control Engineer’s Guide to Amazon AWS – Part 1

The Control Engineer's Guide to AWS - Part 1

It’s not news that Amazon AWS is a growing presence on factory floors across the globe. Many companies now store factory floor data in the AWS cloud, use AWS servers to deploy reporting and monitoring applications, network tools to build scalable and on-demand networks, cloud-based computing services to do analytics, and one or more of the many other services offered by AWS.

But AWS is only one small part of the goliath known as Amazon. This article examines the major components of the AWS portion of the Amazon organization and looks at how AWS came to be and where it fits into that organization.

Before we get into AWS, let’s take a quick look at where it all started: e-commerce.


Amazon e-commerce is where it all started. A simple online bookstore that grew into this juggernaut of a company. How big is the e-commerce portion of Amazon? Is it the overwhelming goliath that its competitors and detractors say it is? Or, as the company says, is it just a large company that’s a relatively small player in retail as a whole? Unfortunately, it’s not as clear cut as you might think.

Amazon is clearly a dominating global retailer. Here’s some unbelievable numbers:

  • Amazon’s revenue in 2021 will be nearly $500 billion dollars! They make nearly $5 million dollars a minute or $79,000 a second. In the time it takes you to read this short article, Amazon will make around a million dollars.
  • More than 4,000 items are sold every minute or about 150 items during the time you read this sentence.
  • They have 150 million Prime subscribers globally. If Amazon Prime users were a country, it would be one of the largest on earth.

But the question is, does it dominate retail or not? Unfortunately, the answer is both yes and no. If you look at overall spending by consumers globally, they do most of their spending, buying the staples of life like housing, food, transportation, healthcare and entertainment, in person. Amazon isn’t much of a player in these areas, which is why the company argues that it isn’t THE dominating global retailer. Not that they aren’t trying to be. Amazon now owns Whole Foods and is working hard to roll out online delivery across the globe. That part of their business is still relatively small but remember, you can quickly go broke betting against Amazon.

The other side of the argument, the one its competitors make, is that Amazon is a massive gorilla overwhelmingly dominating certain retail categories. For example, in books and all kind of electronics, Amazon dominates every category. Is there anyone you know that hasn’t bought a book from Amazon? What about electronics? In these kinds of categories, Amazon has over 50% of sales.

Amazon e-commerce isn’t beloved. It has generated some level of distrust and animosity for the way its warehouse workers have been treated and for other business practices. For example, a recent Wall Street Journal article described how Amazon forced companies that want to do business with it to sell warrants to Amazon. These warrants allow Amazon to purchase large shares in these companies at a future date of their choosing. Not an illegal practice but also not one that generates much goodwill.

Amazon e-commerce is the root of the conglomerate that is Amazon with Amazon AWS cloud now being one of its major branches.


There was a time, 20 years ago, when Amazon wasn’t the tech goliath it is today. Instead, Amazon was just a large e-commerce retail company. It had a well-defined e-commerce engine and a desire to help other large retailers build e-commerce sites on that engine.

But they were facing a situation most of us have faced, projects taking much longer than expected with little commonality to them. Andy Jassy, together with Jeff Bezos, analyzed the situation and found that project teams were building out the same components over and over, components like databases, compute engines and storage components. By creating common components and developing standard APIs for them, Amazon realized they could vastly decrease labor and delivery times for nearly all their project teams.

Realizing the value of these components to scores of other businesses, AWS was born to provide not only databases, compute engines and storage, but a whole host of other standard services like machine learning, human interface and analytics. Organizations from banking to finance to agriculture to logistics are now using AWS cloud services for data storage, backup, email, software development, analytics and customer-facing web applications. Even video game makers now use the cloud to deliver online games to millions of players worldwide.

Cloud services, AWS cloud or a competitor’s, enable the use of a broad range of technologies that were, until recently, much too costly for most businesses. Today, because of the Cloud, a business can obtain storage, compute and database resources at a cost and speed that was unimaginable just a few years ago.

While many companies now offer cloud services, Amazon AWS cloud is the undisputed market leader with approximately one-third of the market. Microsoft Azure has about 20% followed by Google with 7%. With a market growing at 35% in the first quarter of 2021, AWS cloud is destined to build on its success and generate a massive amount of money for Amazon.


With its overwhelming success in cloud computing with Amazon AWS cloud, Amazon next set its sights on IoT (Internet of Things) and created Amazon AWS IoT. AWS IoT is a set of cloud-oriented services for capturing real world data and moving it into its cloud where its cloud services could be used to store, manage, process and display information derived from that data.

AWS IoT sees this process as having three main components: devices to gather data; communications interfaces to move the data; and cloud services to process the data.

Devices are the electronic systems that sense and capture analog and digital data. These devices interface with the real world. They measure and capture all sorts of physical data values, anything from temperature and humidity to river levels, farm crop yields and human faces. These devices are not just input devices. They also can be actuators that use motors for positioning, relays to start and stop systems and human interfaces that communicate with people.

Communication interfaces are the communication technologies and protocols that move data from a device to the cloud. There are hundreds of different types of communication interfaces that are part of the process to move data from the devices to the Cloud, including Wi-Fi, RF, serial, Ethernet, cellular, broadband and more.

Cloud services are the distributed processing, storage and computing services that process the data received from devices over the communication interfaces. This is the area where the AWS cloud connects to the AWS IoT. AWS IoT is a natural extension to the AWS cloud.

What About Manufacturing?

This discussion so far hasn’t included the word “manufacturing” even once. None of this, you should note, has anything specifically to do with automation or manufacturing. That is a key idea behind Amazon AWS IoT that we in manufacturing often miss. The AWS IoT systems were not built to specifically support manufacturing. They were built for more general world applications that are vastly broader and sometimes more impactful than machine data applications. For example, a system that gathers the moisture, temperature and soil properties of an apple grove is an AWS IoT application that has nothing to do with manufacturing, though manufacturing has a subset of applications that are extremely important to AWS.

The next article in this series describes the architecture that implements AWS IoT and will be featured in our November newsletter.