After having started the first class on cloud computing at Stanford, in 2010 I also started a cloud computing class at Tsinghua University in China. The Amazon team was kind enough to give me $3,000 worth of AWS time for the students to use. I showed up in class and told them $3000 would buy a small server in Northern California, Virginia or Ireland for 3.5 years. They looked bored. I went on to say, “Or $3,000 could buy 10,000 servers for 30 minutes.”
This led to the question, “What would you do with 10,000 servers for 30 minutes?”
If you’re an old-timer like me then you know that the last big shift in computing, driven by economics, was the shift to client-server computing. There was a day where PCs and Unix servers were bleeding-edge technology; few thought would be suitable for the enterprise. But the shift happened and companies like Microsoft and Oracle grew to be major enterprise software companies.
As the shift was happening some businesses worked on plans to migrate existing minicomputer or mainframe applications to client-server. While some enterprises succeeded, most failed. Similarly, the major wave to using compute & storage cloud computing will not be driven by migrating existing applications. Instead, as with the move to client/server the driver will be building completely new applications. So what will be these applications? What could be the killer app for cloud computing?
It’s not just about your toaster talking to your coffee maker
I heard about the Internet of Things (IoT) for quite a few years but I mostly ignored it since I wasn’t sure why a toaster would need to talk to a coffee maker. In 2013 I invited Bill Ruh, CEO of GE Digital, to deliver a guest lecture at my Stanford class and his talk piqued my curiosity. It inspired me to learn what was going on in the Industrial Internet, or what some would call enterprise IoT. In order to learn more I reached out to the community and with the help of a crowd of at least one hundred experts, I documented nearly twenty different enterprise IoT case studies spanning all of the major industries: power, water, oil and gas, mining, agriculture, healthcare, construction and transportation. If you’d like to read some of these cases, check out my new book.
The Internet of Things (IoT) is not the Internet of People (IoP)
One of the things I mention in the book is that it’s highly unlikely the software for enterprise or industrial IoT will be the software we have today. All of the software, whether for infrastructure or applications, has been built for the Internet of People (IoP).
Every CRM, ERP, search or ecommerce application is focused on serving people. But things are not people and here are at least five reasons why:
- There are going to be 10-100x more Things connected to the Internet than People.
- Things can be where people aren’t – in your stomach, at the bottom of a coal mine or in the middle of the Australian outback.
- Things have more to say. People can type or click a mouse, but a wind turbine has over 400 sensors on it.
- Things can talk much more frequently. Sensors that help predict a roof collapse for mining machines deliver data at 10,000 times per second, much faster than any of us can point and click.
- And finally, Things can be programmed, while people can’t. So why would software built for the Internet of People work for the Internet of Things?
Some companies have already begun to make the investments in IoT software. GE Digital was founded in 2011 with a one billion dollar investment. CEO Jeff Immelt has declared that GE needed to evolve into a software and analytics company lest its industrial machines become mere commodities. Immelt has set an ambitious target of $15 billion in software revenue by 2020. PTC has invested nearly a billion dollars in a acquiring a series of software companies including ThingWorx, ColdLight and Axeda. On the venture side: Uptake, a Chicago-based IoT startup, beat Slack and Uber to become Forbes 2015’s Hottest Startup.
Is 2017 the year of IoT?
I’ll let you be the judge of whether it’s time to invest in IoT. But if you’re a software developer, mechanical engineer, software excutive, venture capitalist, the CEO of a textile machine company, or the CIO of a hospital, I’d encourage you start learning about the Internet of Things. Consider registering for my new online IoT class, available online, on-demand and in video.