article
What is thin edge?
The “thin edge” is defined by an architecture where data from devices are collected and transmitted for centralized processing. One outcome of an increasingly connected world is an ever-increasing volume of data being generated by machines, devices, and other assets. The sheer volume of this data can represent a major liability or a major advantage to businesses. Leverage data wisely, and you can improve business outcomes via innovations like predictive maintenance. Collapse under the full weight of modern data streams, however, and the economic and logistic consequences can spiral.
Edge architectures—and how they are applied in their different implementations—can be the critical difference between failure and success.
Examples include asset tracking via low-powered sensors—for instance, those providing humidity or vibration monitoring—or other network-centric components. The collected data are then sent to be processed at a location away from the sensors that provided the data in the first place.
Figure 1. IoT at the edge.
Thin vs thick edge
The counterpoint to thin edge is thick edge, where processing is more decentralized, and occurs closer to data collection. An example includes the processing that occurs within a self-driving car. Instead of sending sensor data back to a distant server, which would lead to latencies too high to safely respond to sudden changes on the road, the car’s systems process data directly onboard and autonomously drive the vehicle.
The divide between thin and thick edge, however, is something of a simplification. Most deployments exist on a spectrum between the extremes, making use of varying levels of centralized and decentralized processing as needed.
Figure 2. Thin edge vs. thick edge.
Why use thin edge?
Connecting edge devices to IoT platforms can be hard. And the process is prone to security risks, ecosystem lock-in, burdensome development costs, and issues related to rapid scaling.
As a result, business leaders are empowering product developers to implement thin edge deployments in order to: speed up embedded software development cycles, re-use code, reach deeper into data for analytics, and improve security via encryption, authentication, tamper intrusion protection, and more.
Thin edge allows organizations to deliver edge use cases that need to operate within the constraints of a small memory footprint, low compute power, and anticipated cloud communication issues. Thin edge not only increases the speed and flexibility of organizations implementing IoT projects at the edge but also enables them to leverage the device management features of an IoT platform—including OTA (over-the-air) software management on a broad range of devices, such as programmable logic controllers (PLCs), protocol gateways, and devices.
Thin edge is lightweight and modular, and provides new or current devices with an easy, rapid way to connect to established systems. The benefits of this approach include:
- Being able to add new capabilities to existing edge devices.
- Re-using current apps and making them smart.
- Quickly and easily deploying IoT and Industrial IoT (IIoT) capabilities, despite size, weight, low power, low heat, and other constraints.
In addition, thin edge removes the complexity of keeping software and apps updated and in the field. That’s because its architecture offers strong security and reliability for devices in live operational environments. It also increases speed to market when implemented wisely.
The main characteristics that allow this are:
- Centralized processing—enabling light and low-cost devices and development.
- Cloud or on-premises deployments—improving flexibility.
- Minimal intelligence, data filtering, and aggregation at the edge—leading to analytics transparency and power.
- Potential avoidance of additional gateways—increasing security and saving costs.
Who uses thin edge?
There are numerous use cases for thin edge deployments across many industries and deployments. The following are a few examples.
Municipalities
As municipalities around the globe move to electrify transportation and build out smart infrastructure, thin edge deployments for fleet management have seen record growth. Security, and central organization of vehicle tracking, for instance, are ideal for thin edge devices and computing. Automated alerts can notify response teams to unusual movement or location information, as well as provide ways to offer innovative services, like travel updates or alternative routing, directly to citizens.
Figure 3. Thin edge for municipalities.
Auto parts manufacturing
A smart manufacturer of auto parts depends on thin edge in its industrial PLCs to automate the monitoring of its systems with predictive maintenance, and to reduce downtime, manage operational costs, and improve production quality. Thin edge deployments in such a setting can add sensor data to existing devices or connect modern devices with built-in sensors. Instead of redundant, heavy processing taking place within the production lines themselves, thin edge data is efficiently collected for centralized processing. An added benefit is that a business with multiple lines within a factory, and multiple factories spread across geographic areas, can then use the collected data to identify best practices, compare production costs, efficiency, and quality, and gain better global views of production. Indoor/outdoor asset tracking during and after production represents another use of thin edge for a manufacturer.
Figure 4. Thin edge in manufacturing.
Benefits of thin edge
Thin edge is a powerful tool for:
- Reducing costs and improving customer experience through predictive maintenance.
- Simplifying software management.
- Maintaining connected products at the latest software version.
- Scaling quickly and efficiently.
- Avoiding security issues.
It’s quick, light, low-cost, and highly secure.
Cumulocity through thick and thin
Whether you need real-time data from thick, thin, or any mix along the spectrum of edge computing deployments, Cumulocity’s platform is enterprise grade and built for the future so that you can do more with IoT.
We’ve spent a lot of time thinking about which models of deployment work best in a variety of situations and can help you get started building your Internet of Things, edge solution in minutes. We’re also considered experts at the thin edge—taking top score in MachNation’s in-depth analysis of 12 IoT edge vendors.
So let us show you how to get at the thin edge to do ultra-fast analytics, deploy IoT in new places with limited network connectivity, and cut infrastructure costs.