In a traditional cloud computing model, compute resources and services are often centralized at large datacenters, which are accessed by end users across organizations. This model has proven cost advantages and more efficient resource sharing capabilities. However, new forms of end-user experiences and the use of intelligent AI/ML powered applications are creating a shift to move compute power closer to where a physical device or data source actually exists, i.e. at the network’s "edge."
By placing computing services closer to edge locations or devices, users benefit from faster, more reliable services, while companies benefit by being better able to quickly process data, and support applications without worrying about latency.
Edge devices are physical hardware such as IoT gateways, industrial controllers, smart displays, point of sales terminals, vending machines, robots, and drones. They’re located in remote locations at the edge of the network and are equipped with enough memory, processing power, and computing resources to collect data, process that data, and execute upon it in almost real-time with limited help from other parts of the network. In many cases, organizations can have thousands of edge devices across their architecture, all of which can be managed from a centralized location.
Edge computing can complement a hybrid computing model, and can specifically be used for:
- Several stages of the artificial intelligence/machine learning lifecycle - such as gathering data, deploying apps into production, making inferences and monitoring the operation as new data is collected.
- Coordinating operations across geographies
- Autonomous vehicles
- Augmented reality/virtual reality
- Smart cities
One of the major benefits of edge computing is its ability to optimize resources. If an issue needs to be addressed, only the necessary services and functionality are deployed, which decreases bandwidth usage and costs. Furthermore, if a device loses connection to the core datacenter or cloud, it will continue to operate and maintain remote resiliency.
The Internet of Things (IoT)
The Internet of Things (IoT) refers to the process of connecting everyday physical objects to the internet—from common household objects like lightbulbs; to healthcare assets like medical devices; to wearables, smart devices, and even smart cities.
IoT devices aren’t necessarily edge devices. But these connected devices are part of many organizations’ edge strategies. Edge computing can bring more compute power to the edges of an IoT-enabled network to reduce the latency of communication between IoT-enabled devices and the central IT networks those devices are connected to.
Simply sending or receiving data is what marked the advent of IoT. But sending, receiving, and analyzing data together with IoT applications is a more modern approach made possible by edge computing.
When problems arise in mobile computing, they often revolve around latency issues and service failures. Edge computing can help solve for stringent latency constraints by reducing signal propagation delays. Additionally, it can limit service failures to a smaller area or user population, or provide a degree of service continuity despite intermittent network connectivity.
As service providers modernize their networks, they are moving workloads and services out of the core network (in datacenters) towards the network’s edge: around points of presence and central offices. With the virtualization of central offices, one of the last physical interfaces for service delivery, service providers can reach the goal of deploying services at the network edge.
IoT produces a large amount of data that needs to be processed and analyzed so it can be used. Edge computing moves computing services closer to the end user or the source of the data, such as an IoT device.
Edge computing is a local source of processing and storage for the data and computing needs of IoT devices, which reduces the latency of communication between IoT devices and the central IT networks those devices are connected to.
Edge computing allows you to benefit from the large amount of data created by connected IoT devices. Deploying analytics algorithms and machine learning models to the edge enables data processing to happen locally and be used for rapid decision making.
IIoT stands for Industrial Internet of Things, a term for connected devices in manufacturing, energy, and other industrial practices. IIoT devices are often deployed in connection with edge computing. IIoT is significant for bringing more automation and self-monitoring to industrial machines, helping improve efficiency.
Multi-access edge computing (MEC) is a type of network architecture that provides cloud computing capabilities and an IT service environment at the edge of the network. The goal of MEC is to reduce latency, ensure highly efficient network operation and service delivery, and improve the customer experience.
Multi-access edge computing is now more broadly defined as an evolution in cloud computing that uses mobility, cloud technologies, and edge computing to move application hosts away from a centralized datacenter to the edge of the network, which results in applications that are closer to end users and computing services that are closer to the data created by applications.
5G refers to the fifth generation of mobile networks, representing upgrades in bandwidth and latency that enable services that weren’t possible under older networks. 5G networks promise gigabit speeds—or data transmission speeds of up to 10 Gbps. 5G service also vastly reduces latency and can expand coverage to remote areas.
5G can be considered a use case for edge computing, and it also enables other edge use cases. Edge computing is a way to meet the performance and low latency requirements of 5G networks and improve the customer experience.
Adopting edge computing is a high priority for many telco service providers as they modernize their networks and seek new sources of revenue. Specifically, many service providers are moving workloads and services out of the core network (in datacenters) toward the network’s edge, to points of presence and central offices.
For telcos, the apps and services their customers want to consume on edge networks are the key to revenue generation, but success depends on building the right ecosystem and coordinating among stakeholders and technology partners alike.
No single vendor can provide a complete edge computing solution. Instead, you will assemble a solution from multiple components. Open source platforms ensure interoperability across a wide ecosystem, without the vendor lock-in of a proprietary technology stack. And to enable new edge computing use cases, Red Hat is investing in upstream open source communities like Kubernetes, OpenStack, and Fedora IoT.