IOT technologies are helping decision-makers to quantify new insights from industrial equipment, workers, buildings, and other sources. These real-time edge data insights have the potential to reduce operating costs, improve efficiency, or deliver new capabilities. As businesses increasingly turn to IOT and analytics technologies in order to maximize the efficiency industrial environments, it’s no surprise that Edge Computing has become one of the biggest technology buzzwords of 2017. With massive volumes of data being generated by connected sensors, the Edge Computing model has proven to be an important building block for the industrial analytics applications of the future.
In Part 1 and Part 2 of this series, we reviewed the benefits of deploying Edge Computing in industrial contexts and explored use cases where Edge Computing can deliver the most value. In this final installment of our series on SDX Central’s “Innovations in Edge Computing and MEC” report, we’ll look into key considerations for the future adoption of Edge Computing technologies and discuss a few of the “number of issues that need to be addressed before mainstream adoption” of Edge Computing technologies.
As we discussed in our previous post, “Edge Computing and certain virtual network functions (VNFs) will [be running] in exactly the same location.” Due to this, network operators will need to decide what the optimal infrastructure is to support both Edge Computing deployments and virtual networks. SDX Central recommends that decision-makers considering implementing an Edge Computing solution perform a gap analysis to identify whether the two systems should:
- Share the same infrastructure, or
- Share similar infrastructure, simplifying procurement, or
- Use different infrastructure if the requirements diverge sufficiently e.g. edge requiring GPU etc.
Mobile technologies dominate every aspect of consumer and industrial technologies today. As connected devices are “moving within the network or even when changing radio access technologies (RATs) e.g. WiFi to 5G, users will demand continuity of service.” This is especially important for stateful applications, which must maintain consistency throughout the various nodes of an application. “For stateful application such as AR/VR, there needs to be seamless handoff.” SDX predicts that the consistency could be solved by providing for:
- Mobility of state across edge instances, or
- Mobility of application (VM/container) across edge instances (assuming they are not multi-tenant)
Multi-operator applications, connected cars for example (which each have their own driver-operator), “are typically not all going to be connected to the same communications service providers (CSPs). This means that for car to car or car to sensor communication to occur effectively, the application has to span multiple operators.” Edge Computing helps solve for this problem, by providing a model for P2P communication between edge nodes. Related innovations, such as digital twins, will also aid in smoothing out consistency issues when connectivity is unavailable.
Security and Compliance
When connecting anything to the internet, especially to business systems, security should be a priority. SDX Central states “the edge compute environment has to be secure.” Devices (such as POS terminals) are often the gateway for larger security exploits, therefore it is paramount that security be included from the start. SDX Central explains that “security can be approached from a persona point of view: user, network operator, third-party app provider, app developer, content provider, platform vendor, IoT device. Or it can be approached from an attribute point of view: privacy, integrity, trust, attestation, verification, measurement.” Furthermore, there might be related compliance issues with regard to data storage and security, therefore it is important to only ensure “that only applications authorized to access certain information are able to do so and maintaining data-location as per local regulations.”
According to SDX Central, “edge applications are likely to be very dynamic in nature depending on end user usage patterns.” Because of this, scalability is an important consideration for any application implementing an Edge Computing solution. Edge Computing increases the “surface area” of an application, introducing several new endpoints that must also scale independently. Therefore, “edge compute environments need to scale on multiple axes: performance, capacity, end-users/devices etc.”
Due to potential loss of connectivity, or the need for edge devices to perform control functions in applications that require ultra-low latency, Edge Computing architectures requires a certain degree of autonomy. SDX Central states that “autonomy ensures that edge environments continue to deliver the required services and functionality even when external connectivity or services may be failing. This means that decision making cannot be centralized and the edge environment must have the intelligence it needs in order to continue uninterrupted operation on its own.” This differs from traditional architectures, as independent edge nodes can lose connectivity. With Edge Computing, it is important that the system remains unaffected when a node loses connectivity, and that the individual node can continue operations until connectivity is reestablished.
Operations and Management
“The edge-environment operation will be fully automated given the cost and complexity of manual intervention,” says SDX Central. “For this reason, data collection and closed loop automation will be critical. Furthermore, many of these closed loops may need to be autonomous and not expect connectivity at all times.” When it comes to operations and management, Edge Computing systems must be able to react autonomously, as well as notify human operators when intervention is necessary. However, when connectivity is unavailable, autonomous operations and management features will be responsible for maintaining uptime and efficient performance.
Learn how Swim ESP™ can empower industrial applications by using Edge Computing to optimize and alleviate the sensor data overload. Swim ESP can help transform and manage sensor data using Machine Learning on edge devices.