Making sense of the data deluge with edge computing

The steady proliferation of sophisticated internet-enabled devices, applications and sensors in recent years have transformed both businesses and homes. The Internet of Things (IoT) is triggering an unprecedented data surge which, if harnessed in the correct manner, could redefine how we mine and process information.

Businesses are awash with real-time information but the challenge facing many is how to collate and make sense of it all. Conventional computing architectures premised on a centralized data center fall well short of expectations. Add to it bandwidth constraints, low latency and frequent network disruptions.  

The answer lies in edge computing which shrinks bandwidth costs related to long distance data transmission and addresses the issue of processing real-time applications at the edge. According to the International Data Corporation (IDC), global spending on edge computing is expected to reach $176 billion in 2022, rising 14.8 percent over 2021. It further states that spending by enterprise and service providers on hardware, software and edge solution services will continue to sustain the upward trajectory through 2025, when the figure will nearly touch $274 billion.

What is edge computing?

In simple terms, edge computing refers to the practice of shifting some portion of computation away from the centralized server and bringing it closer to the point where the data is being created. Processing and analysis of the raw data is done at the edge while the findings like real-time insights, predictions and other forms of actionable intelligence is dispatched to the main data center. The principal benefits relate to minimizing latency and cutting down on backhaul traffic volumes and high costs.  

For the telecom sector, edge computing, also known as Mobile Edge Computing (MEC) or Multi-Access Edge Computing provides for application compute and storage close to the end user’s network. The edge infrastructure will typically get managed by communication service providers (CSPs) or other service providers. Here it is crucial to point out that edge computing is not just about MEC but also connectivity and control together with services.

Computing tasks can take multitude of forms and edge computing has emerged as a viable architecture best suited to support distributed computing near the data source. The relevance of edge computing stems from the fact that it could mitigate network congestion due to the movement of vast amounts of data that modern-day businesses produce and devour. It injects speed since the latest cutting-edge applications rely on processing and reactions that are acutely time sensitive.

“To run, edge computing requires server hardware, cloud infrastructure software, cloud stack and the application. The hardware needs to be capable of running the application functionality efficiently while the cloud software enables and controls the virtual infrastructure resources to cloud applications,” says Ismo Matilainen, Marketing Manager, Nokia AirFrame and Edge cloud.

Edge computing is becoming an integral part of complex business operations with the number of devices connected to the internet spiking exponentially and the data volumes produced as a result is going beyond the handling capacity of conventional servers. It is straining the global internet at the same time.  

Taking a case in point, imagine a video camera or a monitor connected to the internet that is sending live footage and data from a factory or office premises. While the network could conveniently manage the transmission at that level, it would be quite a different scenario if there were hundreds or thousands of similar devices performing the same function. It would not only have an impact on quality due to latency but raise bandwidth costs significantly. Edge-computing could be the answer to the problem by processing and storing data locally.

When it comes to the potential locations where telco edge could be deployed, the choices are many, including the premises of the customer, street cabinets, mobile towers and network aggregation points.

Benefits and use cases

Edge computing addresses a plethora of critical infrastructure challenges such as excessive latency, bandwidth limitations and network congestion. But there are a few other advantages that it provides to the end user. For many businesses, the prospect of lowering costs is enough motivation to deploy edge computing, especially when it comes to cloud resources and bandwidth. As time progresses, virtually every household and office premises are expected to install IoT devices and to support such a complex ecosystem, robust computation capabilities will have to be located to the edge.

The technology can be of tremendous help in remote areas with patchy connectivity or poor bandwidth. Edge computing can also keep data secure by doing away with the need of transferring it across large distances, often traversing different regions and international boundaries. By keeping data almost at the source, edge computing offers relief from a multitude of procedural bottlenecks.

The use cases of edge computing spans almost all industry verticals and consumer habits. The value proposition for IoT applications is the most noteworthy. Manufacturing and heavy industries employ it for a variety of factory floor operations, not just limited to monitoring and automated operations of complex machinery. The edge, again, facilitates the integration of IoT applications for predictive analysis and maintenance.  

In retail, minimal latency could be used to create rich, interactive shopping experience in stores or even at home. Retailers often produce a lot of data, ranging from stock tracing, sales and surveillance. Edge computing can help sift through the diverse data and create predictive models and identify novel business opportunities.

Healthcare is set to reap the benefits of edge computing in a major way. Data analysis from IoT devices, sensors and other medical equipment will help swift decision-making and ensure effective medical intervention.

Autonomous vehicles will be generating enormous volumes of data on vehicle, road and traffic conditions, speed and location. The information must be aggregated and studied in real time for safety and reduced journey durations. Such tasks require efficient onboard computing that can only be done at the edge.

Data collected from wide range of sensors and drones from farmlands could provide valuable information on soil temperature, water presence, density of nutrients and predict optimum crop growing and harvesting periods. Edge computing will boost gaming and video experiences by reducing lag time and optimizing high-definition streaming. Data from the radio network can be used to augment network performance. For instance, enabling near real time control (RIC) for beam optimization.

5G and edge computing

The global roll-out of 5G is set to become one of the most defining chapters in the history of the telecoms sector. Promising to be a harbinger of innovative technologies, the latest standard in wireless communications brings with it high bandwidth and ultra-low latency. Driving a new generation of intuitive applications will be software that will run on a distributed architecture along with several components. All finding their place at the network edge.

As carriers get busy rolling out 5G, many of them are fusing edge computing into their deployments. There is considerable overlap between 5G and edge over artificial intelligence and machine learning (AI/ML) Industry 4.0, augmented and virtual reality (AR/VR), automation, self-driving vehicles, IoT etc.

5G will give a fillip to industrial automation. As the functionality on the edge and edge networks grow it would need lot more human resource to handle the network which won’t be cost efficient. “Automation is vital to managing thousands of edge sites, enabling 5G cloud-based CSP offerings and future evolution to new business services at the edge. Edge automation with central management provide the required OPEX savings compared to the current crop of centralized data center, where the number of manual operations is still manageable,” says Matilainen

The future of edge computing

While the initial objectives of edge computing involved lowering bandwidth costs and processing data efficiently, the advent of real-time applications, requiring localized processing and storage will pave the way for further development of the edge. It is imperative that telco operators adopt edge computing if they wish to evolve their networks and discover new revenue sources. The possibilities are truly spectacular, and the technology could prove to be a gamechanger for service providers in their digitalization journey as they adopt the latest fiber technologies and 5G.

About Nokia

We create technology that helps the world act together.

As a trusted partner for critical networks, we are committed to innovation and technology leadership across mobile, fixed and cloud networks. We create value with intellectual property and long-term research, led by the award-winning Nokia Bell Labs.

Adhering to the highest standards of integrity and security, we help build the capabilities needed for a more productive, sustainable and inclusive world.  

Media Inquiries:

Nokia
Communications
Phone: +358 10 448 4900
Email: press.services@nokia.com