This article provides a guide to Edge Computing in 2021. Recently, industry investment and research interest in Edge Computing have grown dramatically. Computing at the Edge moves computing and storage from the Cloud to the Internet’s edge in close proximity to mobile devices or sensors.
Hence, Edge Computing makes it possible to deliver highly responsive services for mobile computing, scalability, and privacy-policy enforcement of Big-data based AI processing at the Edge – Edge AI.
In the following, we will provide an easy-to-understand guide covering the following topics:
- Definition of Edge Computing
- Origin and Advantages of Edge Computing
- Related concepts, Edge vs Fog Computing
What is Edge Computing?
Edge computing is a new computing paradigm that performs computing at the edge of the network. As opposed to cloud computing, edge computing moves closer to the user and closer to the source of data. At the edge of the network, it is lightweight for local, small-scale data storage and processing.
Edge Computing will be 100x what public Cloud is. Michael Dell
With the rapid development of the Internet of Everything (IoE), the number of smart devices connected to the Internet is growing fast, generating large-scale data (Big Data) at the network edge. The massive amount of data causes problems such as bandwidth load, slow response speed, poor security, and poor privacy in traditional cloud computing models.
Hence, traditional cloud computing is no longer sufficient to support the needs of today’s requirements for smart services and intelligent data processing, so edge computing technologies have emerged.
Edge computing directs computational data, applications, and services away from cloud servers to the edge of a network. Application developers can use the on-device computing systems by offering the users services closer to them. Edge computing is characterized in terms of high bandwidth, ultra-low latency, and real-time access to the network information that can be used by several applications.
Therefore, edge computing is the foundation of next-generation Edge Intelligence, the deployment of machine learning algorithms to the edge device where the data is generated.
Edge Computing Definition
Since edge computing has become a research hotspot, there are different definitions. Satyanarayanan, a professor at Carnegie Mellon University, defines Edge Computing as “a new computing model that deploys computing and storage resources (such as cloudlets, micro data centers, or fog nodes, etc.) at the edge of the network closer to mobile devices or sensors”.
China’s Edge Computing Consortium defines Edge Computing as “near the edge of the network or the source of the data, an open platform that integrates core capabilities such as networking, computing, storage, applications, and provides edge intelligent services nearby to meet the industry agility key requirements in connection, real-time business, data optimization, application intelligence, security, and privacy”.
Zha provides the following Edge Computer definition: “Edge computing is a new computing model that unifies resources that are close to the user in geographical distance or network distance to provide computing, storage, and network for applications service”.
Background and Need For Computing at the Edge
Today’s intelligent society is driven by the need for smart and connected services across various industries. Edge devices have spread to all aspects of society, such as smart homes, surveillance cameras, autonomous vehicles, intelligent production robots, and more. As a result, the number of connected devices is constantly increasing.
Therefore, the amount of data generated by devices worldwide is drastically increasing. Based on the continuous and massive growth of data volume and various data processing requirements, cloud-based big data processing has shown many shortcomings:
- Real-time: The transmission of large amounts of data to the cloud results in a large load of network transmission bandwidth, resulting in data transmission delay. Cloud computing will not be able to meet real-time business requirements.
- Energy consumption: The power consumption of data centers has increased significantly. Cloud computing is not able to meet the increasing demand for optimized energy consumption.
- Security and privacy: Uploading the data to the cloud and store them in a centralized environment comes with the risks of privacy leakage or attacks. The probability of data en-route attacks is higher in cloud computing than in edge computing caused by the longer path to the server.
Therefore, the purpose of edge computing technology is to provide services and perform computations at the edge of the network and data generation. The goal of edge computing is to migrate the cloud’s network, computing, storage capabilities, and resources to the edge of the network, and provide intelligent services at the edge.
This is required to meet the critical needs of the IT industry in real-time business, application intelligence, data optimization, security and privacy, and to meet the requirements of low latency and high bandwidth on the network.
Differences of Edge Computing vs Cloud Computing
Traditional cloud computing transfers all data to the cloud computing center through the network and solves the computing and storage problems in a centralized way. However, The emergence of edge computing will not replace cloud computing.
The two concepts complement each other, enabling the digital transformation of industries to a greater extent. Edge computing is an advanced version of cloud computing that reduces latency by bringing the services close to the end-users. It minimizes the load of a cloud by providing resources and services in the Edge network.
In some Internet services, some data needs to be returned to the cloud for processing after being processed by edge computing, such as in-depth analysis of data mining and sharing, which requires the cooperation of both concepts.
The main differences between cloud computing and edge computing are the following:
- Cloud. The main feature of cloud computing is that it can process a large amount of data, conduct in-depth analysis, and plays an important role in non-real-time data processing, such as business decision-making and other fields.
- Edge. On the other hand, edge computing focuses on the local (on-device), and can play a better role in small-scale, real-time intelligent analysis, such as meeting the real-time needs of distributed services.
At this time, edge computing is required to share the pressure of the cloud and take charge of tasks within its scope of the edge. Local data processing reduces the risk of large-scale data loss. Those factors bring stability to services that include connected devices in a network (IoT).
What is Fog Computing?
Fog Computing was introduced by CISCO in 2012 and was defined as “a highly virtualized platform that provides compute, storage, and networking services between end devices and traditional cloud computing Data Centers, typically, but not exclusively located at the edge of the network”.
The OpenFog Consortium definition of Fog Computing is “a system-level horizontal architecture that distributes resources and services of computing, storage, control and networking anywhere along the continuum from Cloud to Things, thereby accelerating the velocity of decision-making”.
Hence, Fog Computing is a concept, where a huge number of heterogeneous, ubiquitous, and geographically distributed devices communicate with the network to perform storage and processing tasks without the intervention of third parties. These tasks can be for supporting basic network functions or new services and applications that run in an isolated environment.
Fog Computing is a smart extension of cloud computing aimed at bridging the gap with IoT devices. Therefore, Fog computing should not be seen as a replacement of the traditional Cloud architecture, but rather as a new architecture that brings IoT, Edge and Cloud computing together.
Edge Computing Characteristics
Edge computing possesses several characteristics similar to cloud computing, but extends the cloud by its specific architecture:
- Geographical Distribution. IoT applications based on sensor networks highly benefit from processing data locally through edge computing platforms. Big data analytics can be performed rapidly with better accuracy. Edge systems enable real-time analytics and AI-processing on a large scale. Examples include sensor networks to monitor the environment, for example, pipeline monitoring or collision avoidance systems.
- Mobility Support. As the number of mobile devices is rapidly growing, Edge computing also supports mobility to communicate directly with mobile devices.
- Location Awareness allows the use of technologies such as GPS to find the location of devices. Location awareness can be used by Edge computing applications such as Fog-based vehicular safety applications and edge-based disaster management.
- Proximity. The availability of the computational resources and services in the local vicinity allows the users to leverage the network context information for making offloading decisions and service usage decisions.
- Low Latency. The low latency of Edge computing enables the users to execute their resource-intensive and delay-sensitive applications on the edge device. Such applications include connected vehicles, remote health monitoring, warehouse logistics, and industrial control systems.
- Heterogeneity refers to the existence of different platforms, architectures, computing, and communication technologies used by the Edge computing elements (end devices, edge servers, and networks).
- Bandwidth intensive use-cases. An increasing amount of data generated by IoT deployments today is bandwidth-intensive, especially the data from video from surveillance cameras. Placing computational resources as close as possible to high-bandwidth data sources implies that much fewer data need to be sent to the distant cloud data centers. For instance, videos and sensor data from hazardous locations can be processed locally to provide real-time information to responders in public safety applications.
Advantages of Edge Computing
Edge computing systems store and process collected data locally on edge devices without uploading it to a cloud computing platform. This comes with several important advantages to lower the pressure of network bandwidth and other disadvantages of cloud computing.
- Performance. Edge computing technology offers fast data processing and analysis, fast response speed enabling real-time services. It provides users with a variety of fast response services, especially in the field of automatic driving intelligent manufacturing, video monitoring, and other location awareness, rapid feedback is especially important. For example, Edge computing enables real-time computer vision applications.
- Privacy and security. The centralized approach of traditional cloud computing requires all data to be uploaded to the cloud for unified processing. Therefore, cloud-based processing exposes significant risks of data loss and data leakage. Edge computing is based on local data, there is no need to upload data to the cloud, which avoids the risks brought by network transmission. Attacks and failures only affect local data, not entire data. On-device computing and AI inference make it possible to guarantee data security. Furthermore, on-device computing enables distributed training of AI models, named Federated Learning.
- Efficiency. On-device computing reduces the amount of data transmitted on the network, reduces the transmission cost and network bandwidth pressure, reduces the energy consumption of local equipment, and improves computing efficiency. Edge computing drives the development of highly efficient AI hardware such as deep learning accelerators (e.g. Vision Processing Units).
- Reliability. Edge computing technology provides methods to make services more stable, robust and highly accessible. The high reliability of connected on-device systems is especially important in mission-critical applications that are vulnerable to network disconnections which may lead to fatal consequences (e.g. medical monitoring or transportation systems).
Edge computing envisions bringing services and utilities of cloud computing closer to the end-user for ensuring fast processing of data-intensive applications.
Learn more about related topics:
- Read about Edge AI – connecting Edge Computing with AI
- An overview on Edge Intelligence, smart services at the Edge
- The concept of Self-supervised Deep Learning
- Learn about Federated Learning to train ML models at the Edge