The case for
private mobile edge
computing (MEC)

The list of potential use cases for mobile edge computing (MEC) is already long and growing by the day. The question for many organizations is not whether to begin using this technology, but whether to do so using public or private networks. This choice, private MEC vs. public MEC, will be influenced by a number of factors including performance, latency, use case, security, control and the data sovereignty they require.

Mobile edge computing (MEC), also known as multi-access edge computing, brings critical computing resources such as processing power and storage closer to end users, devices and applications—at the edge, instead of having to travel to and from a distant data center or cloud.

While MEC can be utilized on 4G LTE or 5G networks, access to remote servers could be more efficient with 5G.

Other benefits of MEC include the potential for:

• Ultra-low latency
• Improved security
• Lowered energy and battery demand in Internet of Things (IoT) devices
• Increased cost efficiencies due to reduced transmission through the cloud
• Increased ability to scale efficiently and quickly

Private mobile edge computing refers to a model of mobile edge computing in which the IT infrastructure—such as compute and storage resources—necessary to support an edge architecture is implemented on-premises for the dedicated use of one specific customer and area, such as an industrial plant or corporate campus. A custom-built private 5G network does not utilize public network infrastructure.

In contrast, public MEC uses edge computing infrastructure built into the public mobile network enabling a pool of shared computing resources available to any customer. This differing level of control over your entire computing and communications stack is one of the major differences between the two options. Public MEC is well suited for organizations that need broad geographic coverage, particularly if they are working with time-sensitive applications that require fast response times.

Private 5G mobile edge computing combines bespoke MEC with a private, custom-built 5G wireless network that is fully managed, monitored and maintained to help organizations solve current IT challenges and gain operational efficiencies. Private MEC is best for companies that require near real-time insights, decision-making and operations, or just additional data sovereignty.

By combining MEC and a private 5G wireless network to create a dedicated, secure edge computing platform with the lowest possible latency to a specific area or endpoint, a private solution offers a number of benefits to users:

• Near real-time performance: Private solutions enable extremely low latency since all of the required resources are on-premises. Critical data doesn't need to leave the site for processing, analytics or any other reason. As a result, applications with low latency and/or high availability requirements are good fits here because the MEC resources are as close as possible.
• Tailor-made: Companies can tailor their services to the specific needs of their customers and business goals. For example, a logistics and supply chain services company could leverage its private 5G network to support computer vision that can help automate and improve its quality control process. This solution requires HD video cameras integrated with a private 5G network, and the company in the example hopes to save between 15% and 30% in potential processing time.
• Data sovereignty: Dedicated on-premises servers enable enterprises to retain control over their data and its availability and security. With a custom-built network, no data leaves the premises unless authorized.
• Enhanced security: The connection between the private network and the existing data center can be integrated through a software-defined WAN (SD WAN), Secure Cloud Interconnect (SCI) or another connectivity service to provide end-to-end security.
• Dedicated computing platform: A private network is for the dedicated use of a single enterprise and location. This enhances performance and reliability above and beyond what's available in multi-tenant, shared resources—a good option for critical applications with compute-intensive requirements that need always-on availability. Most artificial intelligence (AI) disciplines, such as machine learning (ML), natural language processing and computer vision require significant computing resources and massive amounts of data to run optimally.

A private bespoke MEC solution pairs the power of cloud computing with the speed and performance of edge architecture—all locally on-premises. This makes it the choice when security, control and low latency are paramount. Critical and latency-sensitive applications benefit from the ability to not only custom design a network, but better control network performance.

Not all elements within an organization's operations may require this level of security and latency. A private solution can also be used in a hybrid approach with the public 5G network so that devices and applications are distributed across private and public sites, optimizing network value while maintaining the control and security of a private network.

There are many potential use cases:

• Monitor critical infrastructure in near real time, which can reduce inspection costs, improve safety and foster more efficient operations.
• Enhance logistics and supply chain management, as computer vision and ML can ensure better item picking accuracy.
• Utilize augmented, virtual and mixed reality technology on construction sites to help deliver projects on time and on budget and minimize the need for some staff members to travel to the site.
• Improve how machines are used, including via predictive maintenance, based on data analytics from IoT sensors and cameras.
• Automated quality assurance (QA) via low-latency computer vision and ML can stop items with production defects from reaching customers.
• Warehouse robotics and automation, including plant-floor automated guided vehicles, improve operational efficiency, alleviate labor shortages, enhance safety and reduce collisions.