With the advent of the Internet of Things, we’re drowning in data, much of it being delivered to the cloud for ubiquitous accessibility from any device. Data in the cloud expanded exponentially and then created its own set of issues, especially security to identify who should have what level of access. There also remains a latency issue with the cloud. With businesses needing immediate access to data, the cloud can cause delays in delivery.
Enter edge computing, which is strongly on the rise. There is a dire need for this transformation due to the growth of big-data workloads and real-time computing, which has slowed production in the cloud. Edge computing can be defined as an open IT architecture that features decentralized processing power enabling mobile computing and IoT technologies. It generally involves servers located near smart products, with those servers acting as a type of collection point for data computing. In edge computing, data is processed by the device itself or by a local computer rather than being transmitted to a data center/cloud.
That brings other challenges for businesses. How to decide which organizational and operational data needs to be accessed near-term, short-term or long-term (archival) and how to create the appropriate repositories for that data.
Since vast amounts of data can be collected from the billions of devices in the field, the devil is in the details, such as industry-specific regulations and best practices that affect how the company has to implement IoT.
For example, in the oil and gas industry, offshore oil platforms aren’t known to have the best connectivity, yet they process vast amounts of data that tracks rig performance. That data is saved but usually not analyzed in the moment but looked at later. In some cases, with data immediately at hand, valve adjustment in offshore rigs could offer more immediate efficiencies. This is where an edge computing system could be extremely valuable for immediate access to critical data.
IoT is also used in healthcare scenarios, such as with heart and patient monitors, ultrasounds, cancer treatment planning systems and all kinds of smart-enabled medical devices. The value of edge computing in healthcare will become more prominent as the volume and velocity of healthcare data increases at the same rate as the inefficiencies in streaming all this data to a cloud or data center for processing.
This can also be seen when processing real-time purchases at retail outlets using product or facial recognition. The data flow is large, and sending all that data to the cloud and processing it would be too slow and too expensive to work well. However, edge computing is combatting such issues as seen now in the trials of self-service restaurants that are utilizing edge processing to make the purchases effectively.
So what’s the best path to generating the most meaningful results from this digital transformation? The information that can be gleaned from IoT devices – often in real time – deliver new insights for every level of the organization. By disseminating big data and analytics, the potential value IoT brings to any business is vast. There is much to gain, but what’s the best way organizations can identify which benefits will deliver the greatest business results that an IoT transformation strategy is able to achieve?
First identifying what data needs to be accessed immediately and used effectively in remote locations, i.e. at the offshore oil platform, is crucial. As edge computing provides efficient data processed near the source, it can reduce Internet bandwidth usage. This not only reduces costs, it ensures that applications can be accessed effectively in remote locations. It also delivers a level of security and privacy for sensitive data.
The requirements for a successful edge computing transformation include a focus on agility and performance. A single edge device, for example, can’t deliver the computing power or the storage requirements of an enterprise data center. It requires a situation where edge devices create a platform that can absorb data at a speed and volume that doesn’t overwhelm the cloud or a company’s mobile networks. It also must be able to provide user-specific data that can be utilized to make better operational or business decisions, i.e. predicting threats or providing alarms. It should offer real-time situational data, i.e. what’s happening in real time. And it should also include high levels of security.
There are several companies, i.e. HPE’s Aruba, Nutanix, Microsoft Azure, and even Amazon, that are offering applications to help organizations define and launch intelligent platforms at the edge. With an intelligent platform as the basis for an edge computing strategy, the cloud can become a repository for long-term storage, analytics and machine learning. With the aggregation of data from edge devices, the cloud can be utilized to better understand data patterns and trends.
Kevin Meany is Co-Founder and CTO of Versatile.
.svg)
.svg)
