top of page

Edge Computing Explained: How Edge Computing Revolutionizes Real-time Data Processing

Updated: Sep 7, 2023

Volkmar Kunerth, CEO of Accentec Technologies and IoT Business Consultants




Edge Computing Defined


Edge computing refers to computer programs that deliver low latency, are closer to the requests, and involve all computing outside the cloud, at the network's edge, particularly in applications requiring real-time data processing. Edge computing is a distributed computing paradigm that brings computation and data storage closer to data sources, like IoT devices or sensors, instead of relying on a centralized data processing facility. This is done to improve response times and save bandwidth.


In simple terms, edge computing involves processing data at the network's edge, closer to where it is generated, rather than sending it to a centralized data center or cloud for processing. The "edge" can refer to various devices such as computers, IoT devices, or edge servers that are located closer to the data source.


Sensor Edge:


Sensor edge refers to data processing at the source of its generation, i.e., the sensor itself or a device closer to the sensor. Traditionally, sensors collect data and send it to a central location or cloud for processing. However, with the advent of edge computing, data processing is now being done at the edge of the network, closer to the data source. This is crucial for applications that require real-time processing and low latency, such as autonomous vehicles, industrial automation, and smart cities.

Sensor edge computing involves embedding computation capabilities directly into sensors or devices. This allows the processing of data locally, reducing the need to send all the data to a central server or cloud. This reduces the latency, minimizes the bandwidth requirements, and enhances the security of the data.


Combining Sensor Edge and Vision:

Vision, in the context of technology, usually refers to computer vision, a field of computer science that enables computers to interpret and understand the visual world. Computer vision uses algorithms to process images and videos to identify objects, understand scenes, and extract information. This technology is widely used in various applications, such as facial recognition, object detection, and autonomous vehicles.


Combining sensor edge computing and computer vision can lead to powerful applications. For example, in an autonomous vehicle, cameras and sensors continuously capture data about the surrounding environment. This data needs to be processed in real time to make driving decisions. By integrating computer vision algorithms directly into the sensors or devices at the edge, the vehicle can process the data locally and make decisions quickly without sending the data to a central server or cloud.


This combination is also beneficial in industrial automation, where robots and machines need to process visual data in real-time to perform tasks such as object detection, sorting, and assembly. By processing the data at the sensor edge, the machines can operate more efficiently and with lower latency



Device Edge:


Different devices, such as shop floor motors, X-ray machines, and vending machines, are deployed by customers for specific functions. Data from these devices can be analyzed to ensure smooth operations and predict maintenance needs. Computing resources are placed near the devices to process workloads and deliver low-latency responses. Small appliances and gateways commonly provide compute and physical connections to legacy interfaces.


Extended Discussion:


Deploying various devices for specific functions is a common practice across different industries. For example, shop floor motors are essential in manufacturing, X-ray machines are crucial in healthcare, and vending machines are widespread in public spaces and offices. These devices continuously generate data related to their operations, performance, and status.


Ensuring Smooth Operations:

Analyzing these devices' data in real time ensures smooth operations. For instance, the data from shop floor motors can provide insights into their performance, efficiency, and any deviations from the expected parameters. Similarly, data from X-ray machines can indicate the quality of the images being produced, the exposure levels, and the overall functioning of the machine. Analyzing this data in real-time can help identify any issues immediately and take corrective actions to prevent disruptions in operations.


Predicting Maintenance Needs:

Analyzing the data from these devices can also help in predicting maintenance needs. For example, suppose the data from a shop floor motor indicates that it is consistently operating at a higher temperature than usual. In that case, it may be a sign that the motor needs maintenance. Similarly, if the data from a vending machine indicates that a particular item is being dispensed more frequently, it may indicate that the stock needs to be replenished. Organizations can proactively plan maintenance activities by analyzing the data and predicting maintenance needs, minimizing downtime and ensuring continuous operations.


Low-Latency Responses:

Placing computing resources near the devices is essential for processing workloads and delivering low-latency responses. For example, in a manufacturing setup, the data from shop floor motors must be processed in real-time to ensure that the production line is operating efficiently. Any delay in processing the data and responding to any issues can lead to disruptions in the production process. Therefore, computing resources are placed near the devices to ensure low-latency responses.


Small Appliances and Gateways:

Small appliances and gateways are commonly used to provide compute and physical connections to legacy interfaces. These appliances and gateways are equipped with the necessary computing power and connectivity options to process the data from the devices and send it to the central server or cloud for further analysis. They also provide physical connections to legacy interfaces, ensuring that the devices can be integrated into the existing infrastructure without significant modifications.


Router Edge:

Routers primarily forward packets between networks and act as the boundary between external systems and internal networks. Some enterprise routers have built-in computing or can accommodate additional compute modules to host applications. A single router can perform packet routing functions in this model and provide infrastructure to host edge applications. Router edge computing refers to directly integrating computing resources and capabilities into network routers, enabling data processing at the network's edge, closer to the data source. This subset of edge computing involves bringing computing closer to the data source, whether it's an IoT device, a sensor, or any other device generating data.


Components:

Router: A router is a networking device that forwards data packets between computer networks. Routers perform the traffic-directing functions on the Internet. A packet is typically forwarded through the networks from one router to another until it reaches its destination node.

Computing Resources: These include processors, memory, and storage integrated into the router to enable data processing at the network's edge.

Functioning:

Data Processing: The integrated computing resources in the router enable it to process data locally instead of sending all the data to a central server or cloud for processing. This can include tasks such as filtering, aggregation, and data analysis

Traffic Routing: The router continues to perform its primary function of directing data packets between different networks. However, with the added computing resources, it can also make intelligent decisions about data routing based on the processed information. For example, it can prioritize certain data types or optimize the routing path for lower latency.

Advantages:

Reduced Latency: By processing data locally at the router, the time it takes for the data to travel back and forth between the data source and the central server or cloud is reduced. This is crucial for applications that require real-time processing and decision-making.

Bandwidth Savings: Processing data locally at the router reduces the amount of data that needs to be transmitted over the network, leading to significant bandwidth savings.

Improved Security: Processing data locally can minimize the exposure of sensitive data as it does not need to be transmitted over the network.

Scalability: Router edge computing can lead to better scalability as the number of devices and data generated increases. The data processing load is distributed across multiple routers at the network's edge, reducing the load on the central server or cloud.


Applications:


IoT Devices: Many IoT devices generate a large amount of data that needs to be processed in real-time. Router edge computing can enable the processing of this data locally, leading to lower latency and bandwidth savings.

Content Delivery: Content delivery networks (CDNs) can benefit from router edge computing by caching content closer to the end-users, leading to faster content delivery and a better user experience.

Real-time Analytics: Applications that require real-time analytics, such as traffic monitoring and emergency response systems, can benefit from the low latency and real-time processing capabilities of router edge computing.


Branch Edge:


A branch is a location other than the main office designated for specific functions. Each branch uses various applications for its daily operations. For example, a retail clinic may use a Point of Sale system, while a health clinic may use an Electronic Medical Record. Hosting such critical applications on edge computing at the branch ensures low-latency access and business continuity. Edge computing appliances at the branch typically have more capacity than other edge computing servers. They can host multiple virtual network functions and applications on the same hardware. "Branch edge" and "Local Area Network Edge" are sometimes interchangeable.


Enterprise Edge:


In a distributed enterprise environment with many branches, computing resources can be shared among branches to achieve economies of scale and simplify management. Instead of deploying edge computing instances at each location, resources can be implemented at a shared site connected to the enterprise network. This model offers higher capacity and capabilities for applications requiring more processing power and resources.


Datacenter Edge:


As customers move to the cloud from existing data centers, smaller data center variants have emerged for rapid deployment, portability, special events, and disaster management. These can be deployed closer to the customer and vary in size from a suitcase to a shipping container.


Cloud Edge:


Cloud service providers have developed services for specific purposes closer to users to optimize functions like content delivery. While some loosely refer to Content Delivery Networks (CDN) and caching services as cloud edge, they were not designed for general-purpose workloads. Although initial efforts focused on caching and content delivery, newer services like local zones have redefined cloud edge. Cloud service providers have also developed many edge solutions fitting into some previously discussed models.


Mobile Edge:


Wireless service providers offer nationwide service using a highly distributed network. Service locations are usually closer to customers than cloud data centers. When these locations serve dual purposes, providing wireless and hosting edge computing services creates a unique edge computing model with distinct advantages. In mobile edge computing, computing resources are deployed at service access points (SAP) or other core locations. Applications on these edge computing servers can be accessed from mobile endpoints through 4G or 5G connections.


Mobile Edge Computing


Wireless service providers offer nationwide service using a highly distributed network. Service locations are usually closer to customers than cloud data centers. When these locations serve dual purposes, providing wireless and hosting edge computing services creates a unique edge computing model with distinct advantages.


In mobile edge computing, computing resources are deployed at service access points (SAP) or other core locations. Applications on these edge computing servers can be accessed from mobile endpoints through 4G or 5G connections.


Dual Purpose Locations:


Service and Hosting: Service locations traditionally provide wireless services to customers. However, when these locations also host edge computing services, they serve a dual purpose. This dual functionality creates a unique edge computing model that brings several advantages, including reduced latency, bandwidth savings, and improved user experience.

Deployment at Service Access Points:


Computing Resources: In mobile edge computing, computing resources such as servers, storage, and networking equipment are deployed at service access points (SAP) or other core locations in the network. These locations are closer to the end-users, enabling low-latency access to applications and services.

Access from Mobile Endpoints:


4G and 5G Connections: Applications hosted on the edge computing servers can be accessed from mobile endpoints, such as smartphones, tablets, and IoT devices, through 4G or 5G connections. This enables high-speed access to applications and services, even for users on the move.


Real-World Examples of Edge Computing


Autonomous Vehicles:

Example: Self-driving cars generate and process vast amounts of data in real-time to make driving decisions.

Application: Edge computing allows these vehicles to process data locally, reducing latency and enabling real-time decision-making, which is crucial for safe driving.


Smart Cities:

Example: Traffic management systems in smart cities collect and analyze data from various sources such as cameras, sensors, and GPS devices to optimize traffic flow and reduce congestion.

Application: Edge computing enables the processing of this data closer to the source, leading to faster response times and more efficient traffic management.


Industrial Automation:

Example: In a manufacturing plant, sensors areal timend machines continuously generate data on their operations, performance, and status.

Application: Edge computing allows for the processing of this data locally, enabling real-time monitoring and control of the machines, predictive maintenance, and optimization of the production process.


Healthcare:

Example: Wearable devices and sensors monitor patients' vital signs and other health-related data in real-time.

Application: Edge computing enables the processing of this data locally, allowing for real-time monitoring of patient's health and timely intervention if necessary.


Retail:

Example: Retail stores use cameras and sensors to monitor customer behavior, inventory levels, and store conditions.

Application: Edge computing allows for the processing of this data locally, enabling real-time insights that can be used to optimize store layout, inventory management, and customer experience.

Content Delivery:

Example: Content delivery networks (CDNs) distribute content across various locations to deliver it more quickly to end-users.

Application: Edge computing enables the caching of content closer to the end-users, leading to faster content delivery and a better user experience.


Augmented Reality (AR) and Virtual Reality (VR):

Example: AR and VR applications require the processing of large amounts of data in real-time to render images and provide a smooth user experience.

Application: Edge computing enables the processing of this data closer to the source, reducing latency and enabling a more immersive and responsive AR/VR experience.

Internet of Things (IoT):


Example: Smart home devices such as thermostats, security cameras, and voice assistants generate and process data for various services.

Application: Edge computing enables the processing of this data locally, leading to faster response times and more efficient operation of the devices.


Sources:



Volkmar Kunerth CEO Accentec Technologies LLC & IoT Business Consultants Email: kunerth@accentectechnologies.com Website: www.accentectechnologies.com | www.iotbusinessconsultants.com Phone: +1 (650) 814-3266

Schedule a meeting with me on Calendly: 15-min slot

Check out our latest content on YouTube

Subscribe to my Newsletter, IoT & Beyond, on LinkedIn.







Comentarios


bottom of page