We believe device makers and their customers must understand and address five principles to successfully navigate a world made up of billions of connections between people, devices, organizations, and ideas.
The breathtaking proliferation of devices in business today is why we have identified unbounded scale as our second principle, which is inextricably linked to the unprecedented speed we are also experiencing. You cannot compromise speed as you grow.
Why has scale become so relevant so quickly? Think of all the technological advances that are spurring the use of connected devices.
The cloud is a huge enabler, providing essentially an infinite amount of flexible capacity (if you are willing to pay for it, of course.) The dawn of 5G is about more than speed, it also allows devices to work with mobile connections, no longer limited to a wired experience or brick-and-mortar settings. And individual devices have become cheaper, making it affordable to equip workers with, say, handheld machines to manage warehouse inventory or iPads to check in airline passengers.
To me, the essence of unbounded scale is the ability to look around corners and anticipate what the future will look like. When a company initially launches connected devices, you don’t know what you don’t know. The first one hundred devices may be in the field and working well, but when you get to one thousand, new issues arise. You must be thinking three steps ahead.
We tend to think of scale in terms of volume, the sheer numbers of identical devices, such as a million smart meters tracking the electricity used in private homes. This requires more sophisticated ways to process and store the incredible volume of data collected in these systems as well. For example, a grocery store could include one hundred freezers, with each freezer churning out multiple sensor readings every few minutes. Across thousands of stores, this results in a huge quantity of data.
Scale also produces complexity. We increasingly see places where more machines mean diverse devices that need to talk to each other — think of the dozens of microprocessors in your car.
This is especially true in a hospital setting, where multiple systems are managing one patient. Perhaps one device is monitoring vital signs, another is watching brain activity, and multiple infusion pumps are coordinating several medications. As an added complication, individual departments in hospitals often order specialized imaging equipment — say an x-ray device for spinal imaging or an MRI machine for a shoulder injury – without consulting IT, creating new levels of complexity in the system.
Scale becomes a problem if you are not designed for the constant ability to adapt scale in the face of unplanned growth. It can start out innocuously enough with a few devices and then a few more, and then suddenly you’re connecting an entire system of hundreds of units to another system, a stepwise increase both in the number and complexity of the devices. If a system outgrows the original parameters and you must rationalize a solution after the fact, you really don’t have much of a prayer.
Scale, of course, impacts security. The more machines, the easier it is to find a way in, and the damage done can be devastating. Here again, scale is closely linked with speed. The old IT models no longer work when a problem strikes. Humans are not fast enough to assess and react at scale, so you must use machine learning and other elements of artificial intelligence to make devices do this for you.
Scale also offers opportunity. As you add devices, you get better data and better insights from a richer understanding of how customers are using the devices. Scale makes you smarter.
We will continue to address the remaining principles in future blogs: always on, systemic learning, and collective wisdom.