Edge Computing VS Cloud Computing

We are all familiar with cloud computing, which has many features: Huge computing power, massive storage capacity, through different software tools, you can build a variety of applications, we are using a lot of APP, essentially rely on a variety of cloud computing technology, such as live video platforms, e-commerce platforms. Edge computing is born out of cloud computing, close to the device side, with rapid response capabilities, but can not cope with a large number of computing and storage occasions.

The concept of edge computing is relative to cloud computing, cloud computing processing is to upload all the data to the cloud data centre or server processing of the centralized computing resources, any need to access the information request must be uploaded to the cloud processing. Therefore, the disadvantages of cloud computing in the face of the era of the explosion of the volume of IoT data have gradually come to the fore:

Cloud computing can not meet the exploding massive data processing requirements. With the integration of the Internet and various industries, especially after the popularity of the Internet of Things technology, computing demand has exploded, the traditional cloud computing architecture will not be able to meet such a huge demand for computing.

Cloud computing can not meet the demand for real-time data processing. Under the traditional cloud computing model, IoT data is collected by the terminal to be transmitted to the cloud computing centre, and then returned to the results through the cluster computing, which is bound to appear longer response time, but some emerging application scenarios such as unmanned driving, intelligent mining, etc., the response time has extremely high requirements, relying on cloud computing is not realistic.

The emergence of edge computing can, to a certain extent, solve these problems encountered in cloud computing. As shown in the figure below, the data generated by the IOT terminal equipment does not need to be transmitted to a distant cloud data centre processing, but rather close to the edge side of the network to complete the data analysis and processing, compared with cloud computing is more efficient and secure.

Having said that, summarise the advantages of edge computing:

Low Latency: Computing power is deployed near the device side, real-time response to device requests; for example: in the field of face recognition, the response time is reduced from 900ms to 169ms; voice recognition function if processed by the cloud, we get the output latency on the terminal can be perceived, the speed is slower because of the long-distance transmission of network signals. And if localised processing is carried out without network transmission, the delay will be greatly reduced and the user experience will be better.

Low Bandwidth Operation: The ability to migrate work closer to the user or data collection endpoint can reduce the impact of site bandwidth constraints. This is especially true if the edge node service reduces the need to send large amounts of data processing requests to the hub.

Reduced Energy Consumption: For a given task, a decision needs to be made as to whether it is more resource efficient to compute locally or to transmit the computation to other nodes. If the local area is idle, then of course it is the most resource-efficient to compute locally; if the local area is busy, then it is more appropriate to distribute the computation task to other nodes. It is important to weigh the energy consumed by computation against the energy consumed by network transmission. Generally when the resources consumed by network transmission are much smaller than the energy consumed by computing locally, we will consider using edge computing to offload computing tasks to other idle nodes to help achieve load balancing and ensure high performance of each node.

Privacy Protection: Data is collected locally, analysed locally, and processed locally, which effectively reduces the chances of data being exposed to the public network and protects data privacy. For example, we are familiar with the privacy and security features on mobile phones, which can be unlocked by fingerprint recognition and facial recognition, which also use edge computing. If this data is uploaded to the cloud, we risk facing data transparency, so edge computing is better for users to protect their privacy and security.

X

Contact Us

Contact Us