Edge Computing is Imperative

In a recent session at the IoT M2M Council (IMC) Winter IoT Days, Aeris CTO, Syed ‘Z’ Hosain, discussed the role of edge computing in mitigating the data processing challenges in the IoT ecosystem.

Edge computing is the first step in the IoT data chain, enabling the initial processing of data before it is transmitted to application servers for additional analysis. Earlier generations of IoT solutions simply collected data and sent that raw, unprocessed data to the server. While this practice is still sufficient today for applications such as those that, say, take a volume reading on a fuel tank once a day and alert the end user when the levels are getting low, it is simply not able to support applications that require higher bandwidth or lower latency. 

Consider a video telematics solution for fleet operators that records external footage in case of an accident and also monitors the driver’s alertness. This solution will run into two problems in the absence of edge computing: without the intelligence to recognize an accident, the device will have to transmit a large amount of (ultimately unimportant) data at a high cost, and without the ability to recognize an inattentive driver, the device will not be able to provide an immediate alert to correct the behavior. Additionally, such solutions may encounter situations where data must be processed immediately, but the cellular signal has been lost. To deliver on its value, this solution requires a means to minimize unnecessary data transmission, reduce latency, and process data locally in case of a dropped signal. 

Enable Security at the Edge

As IoT solutions become a larger part of our lives, the network of devices hackers can access for malicious purposes grows. Over the last several years, there has been a significant increase in the scale of these attacks. For instance, in 2016, the largest distributed denial of service (DDoS) attack brought down sites such as Twitter, Netflix, CNN, and many more. It achieved this by taking over a large number of connected video cameras, DVRs, and the like, and using those devices to bombard the largest domain name system (DNS) server with traffic until it collapsed.

Without proper security controls, our personal information and safety also becomes much more vulnerable to cyber attacks. Cardiac devices that monitor and control patients’ heart function could be hacked and controlled by malicious actors; baby sleep monitors could be accessed by hackers to snoop on your child; your car could be taken over and controlled from afar. (Yes, these are all real vulnerabilities that were discovered in products on the market.) These examples all point to the crucial need for security at the edge. While cloud servers have robust security measures in place, the data must be secured as early in the IoT data chain as possible to ensure the solution is protected from breaches. Password management, support for over-the-air firmware updates, and encryption on local Wi-Fi networks and cellular gateways are critical components of an end-to-end security protocol for an IoT solution.

Don’t Build a “Data Museum”

If all those reasons aren’t enough justification for building smarter IoT devices, consider the long-term value that IoT can provide to the world. As the number of connected devices doubles over the next several years and those devices send more data, we are at risk of creating a “data museum”, ie, a place where data is stored, collects dust, and loses value. Only analyzed data provides value. But if the back-end servers become so overloaded with data from tens of billions of devices, it may become too costly to send, process, and search that data, effectively rendering it useless. Smarter processing at the edge controls the amount and quality of data that is transmitted to cloud servers, ensuring that the volume of data in the cloud does not overwhelm our ability to process and analyze it.

Whether we reach the low-end estimates of 25 billion connected devices by 2025 or the high-end estimates of 56 billion, it’s clear that IoT solution providers need to prepare for the explosive growth by enhancing edge computing capabilities. Building smarter IoT devices will:

• Reduce costs and improve performance when operating at scale
• Enable tighter security throughout the IoT data chain
• Ensure the value and impact delivered by IoT solutions grows with the scale of deployments

Let Aeris Help

Aeris helps businesses deliver smart and secure IoT solutions. The Aeris Intelligent IoT Network offers “secure by design” cellular connectivity, bringing best-in-class security—ranging from tight, role-based access control to password management—right to the edge. The Aeris Intelligent IoT Network’s security capabilities include:

• Non-dialable number assignments
• The ability to restrict connections to only authorized endpoints
• Secure API access to monitor, manage, and control devices at scale
• A direct and private connection to the cloud

Intelligent also has deep integrations with the underlying carrier networks to enable unparalleled visibility and control into device/network interactions, and is backed by a team of IoT experts who can assist with configuration and testing to enable smarter solutions. For instance, we helped one European fleet video telematics solution provider solve a configuration issue that was causing unacceptably long upload times. We applied our best practices to reduce latency by 70% and enable them to successfully launch their solution.

The Aeris Intelligent IoT Network is the only cellular IoT network that provides visibility and control over your entire connected operation – globally and at scale. With non-steered coverage from 600 carriers in 190 countries, we help customers manage risks and deliver the highest possible quality of service – all while optimizing costs across the business.

Contact us to learn how the Aeris Intelligent IoT Network can help you operate smarter IoT devices at the edge.