The rise of edge computing is critical to scale up IoT deployment, owing to its reduced bandwidth and latency, plus faster application response times.
Mission-critical applications such as factory automation require not only ultra-low latency but also high reliability and fast, on-the-fly decision-making.
Conventional centralized communication architectures are not able to provide the new performance requirements mostly due to congestion, high latency and slow backhaul links.
Furthermore, fast decision-making on highly automated machinery needs advanced computing capabilities right on the spot, which can be provided only by onboard computers or interconnected edge-computing local nodes working together.
Edge computing speaks to a computing topology that places content, computing and processing closer to the user/things or “edge” of the networking. It is not a competing approach to cloud computing but a complementary one.
As mentioned in an earlier article, by 2020, we’ll be reaching Phase 5 of industrial IoT adoption, when businesses recognize IoT as not just a technology but a new business transformation tool that requires rethinking industrial processes to achieve its full potential.
By then, traditional industries will begin the transformation to full digital businesses. Edge-computing services will drive this transformation and accelerate the ROI of deployment of IoT because it will allow new applications to optimize industrial processes.
Whereas cloud computing relies on a limited number of centralized data centers, adding processing power to local devices within the network reduces reliance on the cloud and can distribute processing across a large number of edge nodes. Edge computing provides the IoT with:
- Low latency: Because no cloud round-trip is required for processing data, the response time is significantly reduced.
- Reduced bandwidth: A local network of edge nodes doesn’t need to send all data to the internet, thus reducing bandwidth requirements.
- Further reliability: Edge devices, with their own data processing capabilities, will likely continue to perform most of their functions in cases of lost connectivity to the data center.
- Increased security: As not all data is sent to the cloud, protecting critical, sensitive information can be assured by local encryption. Also, edge devices can be isolated within their local network, preventing external attacks.
Smart Wi-Fi access points and routers can become data collectors and edge-computing nodes for a large number of IoT devices. Current models of enterprise-grade wireless communications devices have processing capabilities that can be exploited for basic edge-computing tasks.
Furthermore, the high-density of mobile devices provides additional processing power, reducing interference and latency.
“The edge-computing structure reduces the network traffic from IoT devices to cloud servers [because] edge nodes upload reduced intermediate data instead of input data,” says Dr. He Li from the Muroran Institute of Technology in Japan. “[Because] only the intermediate data or results need to be transferred from the devices to the cloud service, the pressure on the network is relieved with less transferring data,” he says in his paper, “Learning IoT in Edge: Deep Learning for the Internet of Things with Edge Computing.”
Technologies such as robust dynamic network architecture (RDNA), combined with edge computing, can leverage the latest advances of mobile devices to provide low-cost, ubiquitous communications and computing.
When industries combine the power of IoT with edge computing, they’ll begin to realize the potential of the technology for new, robust applications. Then when AI kicks in as the real game-changer, we’ll see the full industrial revolution—the so-called “Industrie 4.0”—unfold.
