Every consumer who has ever checked a movie schedule or restaurant review with their smartphone knows the benefits of cloud computing. But is the cloud computing model adequate for the industrial world?

Unlike consumer or enterprise deployments, industrial applications cannot rely exclusively on centralized data storage and processing. These apps frequently need immediate action, unencumbered by transmission bottlenecks related to truly remote locations. Real-time response requires an edge-plus-cloud architecture for the 21st century Industrial Internet.

In the mission-critical industrial world, edge devices operate in some of the most difficult terrain on earth. The plummeting cost of sensing technologies allows the collection of massive data sets from these remote and inaccessible systems.

Typical examples of safety-critical industrial machines, at the far end of a long and tenuous link to a central server, may include:

• Water injection pumps battling 50°C temperatures in the middle of the desert;
• Undersea blowout preventers deployed more than 10,000 feet beneath offshore oil rigs;
• Aircraft engines providing 94,000 pounds of engine thrust while flying miles above the earth

Responding In Real-time Is the ‘Edge’ In Edge Computing

The harsh and remote conditions of many industrial applications challenge their ability to connect and cost-effectively transmit large quantities of data in real-time. But what if instead of collecting data for transmission to the cloud, these industrial applications could also process, analyze and act upon the collected data? What if you can add intelligence to machines at the edge of the network, in the plant or field, and in the cloud?

What I want to suggest is that ‘edge’ computing, working in conjunction with the cloud, is a strategic architectural choice that is essentially the front door to the data kingdom, and a modern Industrial Internet platform is a key that permits entry. This encapsulates the true potential of the Industrial Internet, providing outcomes that matter.

But, what is edge?

Edge is the physical location that allows computing closer to the source of data.

Edge Computing enables analytics and resulting insights to occur closer to the deployed machines (i.e., the sources of the data). It supports computing at the edges of a network.

Instead of simply storing data and missing the opportunity to capitalize on it, edge devices can analyze data to gain insights then act on them. Now real-time analytics and application logic can run at various levels: at the edge sensor, controller, gateway, infrastructure machine, within on-premise appliances and racks, or in the cloud.

This edge-to-cloud continuum can be optimized to run at the individual asset level or at the fleet or plant level. In this world, isn’t edge the entry point to the data kingdom? At a minimum, it is the gateway for optimizing industrial data. It provides data and analytic insights that optimize every aspect of asset and physical operations.

In the mission-critical industrial world, edge devices operate in some of the most difficult terrain on earth. The plummeting cost of sensing technologies allows the collection of massive data sets from these remote and inaccessible systems. 

Edge-based software-defined machines (SDMs) like blowout preventers, turbines, medical equipment, streetlights, and locomotives leverage edge-to-cloud architecture. These SDMs are evolving and leveraging the modern edge-to-cloud platform for the Industrial Internet. Now, applications can not only collect data locally and respond to changes in that data, but they can also perform meaningful localized analytics.

Safer, More Efficient Locomotives

The locomotive, the ultimate symbol of the first industrial revolution (19th century), is today a pulsing, data-centric example of 21st century edge-to-cloud computing.

A modern locomotive such as GE Transportation’s Evolution Series Tier 4 Locomotive, carries more than 200 sensors that collect gigabytes of information, processing over one billion instructions per second.  The Tier 4 uses on-board edge computing to analyze data and apply algorithms for running smarter and more efficiently.

It’s one of the smartest locomotives on the planet, and directly improves the locomotive operator’s bottom line.

 GE wants to connect all of its 21,000 locomotives, pulling freight and passengers in 50 countries, to Predix, its platform for the Industrial Internet. 

These Translate Into Multi-billion Dollar Savings For Customers

Even with incredibly intelligent and reliable locomotives, railroads still experience hundreds of thousands of unexpected operational delays each year, costing the industry billions. The latest generation of trains can use edge and cloud computing to facilitate smarter maintenance, longer uptime, and creation of game-changing equipment.

Today’s locomotives are generating more data than ever –  edge and cloud computing provide a smarter way to retrieve and manage data.  The on-board equipment works in conjunction with centralized servers that host policy rules.

Similar Benefits Cut Across Multiple Industrial Sectors

“The edge-to-cloud approach is rapidly gaining favor throughout the industrial sector,” according to Greg Gorbach, vice president, ARC Advisory Group. “It’s a way to deploy powerful analytics where they are needed, while enhancing real-time performance optimization and minimizing data bottlenecks.”

So as you begin evaluating the deployment options for your Industrial Internet solution, I suggest a few probing questions:

• How could edge data be leveraged to enhance your customer experience?
• What value could be generated if connected edge devices have the necessary compute power?
• Do you have a need to optimize aspects within your physical operations and integrate it with other business computing platforms?
• Is there an outcome that could be served by providing real-time data processing and edge analytics?
• And, how could you best leverage the modern cloud and edge architecture for industrial-scale IoT?

Edge computing is not new. The main question will be, “Does it create a door leading to the value of customer data in Industrial applications, just as the smartphone opened the door leading to value of consumer data in mobile apps?”