The data generated at the edge of the network is gradually increasing and if we can process and analyse the data at the edge nodes of the network, then this computing model will be more efficient. Many new computing models are being proposed as we find that with the growth of IoT, cloud computing is not always as efficient as it should be.
1. The Cloud Services Are Driving It:
Cloud centres have powerful processing performance and are capable of handling huge amounts of data. However, transferring huge amount of data to the cloud centre becomes a challenge. The system performance bottleneck of the cloud computing model is the limited network bandwidth, it takes a certain amount of time to transmit massive data, and the cloud centre needs a certain amount of time to process the data, which increases the request response time and the user experience is extremely poor.
2. Driven by the Internet of Things (IoT)
The rapid development of IoT technology makes more and more ordinary objects with independent functions interconnect and achieve the interconnection of everything. Thanks to the characteristics of the Internet of Things, all industries are using IoT technology to rapidly achieve digital transformation, and more and more end devices in the industry are connected through the network.
However, the Internet of Things as a huge and complex system, different industries have different application scenarios, according to third-party analysts, by 2025 there will be more than 100 billion terminal devices connected to the network, the terminal data volume will be up to 300ZB, such a large-scale data volume, according to the traditional data processing, all the data obtained need to be sent to the cloud computing platform to analyse, the cloud computing platform will be faced with high network latency, mass access to devices, and difficult to process large amounts of data. The cloud computing platform will face the challenges of high network latency, massive equipment access, massive data processing, insufficient bandwidth and high power consumption.
In order to solve the drawbacks of high latency and lack of real-time data analysis capability in traditional data processing methods, edge computing technology has emerged. Edge computing technology is in close proximity to the object or data source on the edge side of the network, through the integration of network, computing, storage, application core capabilities of the distributed open platform, close to the edge to provide intelligent services. To put it simply, edge computing is to take the data collected from the terminal and analyse it directly and aggressively in the local device or network close to where the data is generated, eliminating the need to transmit the data to a cloud-based data processing centre.
For example, the real-time operational and safety concerns required for self-driving cars are pushing the computing core from the cloud to the edge of the network. Self-driving vehicles are constantly sensing and sending data about road conditions, location and surrounding vehicles. Self-driving cars generate about 1 GB of data per second, and the processing bandwidth and latency required makes it impractical to send even a fraction of a terabyte (TB) of data to a centralised server for analysis. Processing data quickly is a critical capability, and edge computing is key to enabling autonomous driving. For vehicles to operate safely and reliably, any lag in processing speed could be fatal.
Imagine a self-driving car detecting objects on the road, or operating the brakes or steering wheel with delays due to the cloud. Any slowdown in data processing will result in a slower response from the vehicle. If the slower responding vehicle is unable to react in a timely manner, it could lead to an accident. Lives can be effectively threatened at this point.
Therefore, it is necessary to provide enough computing power and reasonable energy consumption to ensure the safety of self-driving vehicles even at high speeds. The primary challenges in designing an edge computing ecosystem for self-driving vehicles are to provide real-time processing, sufficient computing power, reliability, scalability, cost, and security to ensure the safety and quality of the user experience in self-driving vehicles.