Alibaba Cloud View Computing, the main "battle" field of edge computing

Introduction to Cloud Computing Intelligence Bureau, Alibaba Cloud product expert Yun Jue conducted a comprehensive online demystification of the product design background, product functions, application scenarios and value of the new product-View Computing, leading netizens to explore the new "vision" "Boundary.

\>>Conference Portal : https://yqh.aliyun.com/live/edgecloud\_visual

Recently, in the 10th issue of the Cloud Computing Intelligence Bureau, Alibaba Cloud product expert Yun Jue conducted a comprehensive online reveal on the product design background, product functions, application scenarios and value of the new product-View Computing, leading netizens to explore a new "view" world.

In the 5G era, view data has become the most important carrier of information digitization

In Yunjue's view, the large-scale digitization of information can be roughly divided into three stages: the first stage is the digitization of text, the second stage is the digitization of voice, and the third stage is the large-scale digitization of video and picture data.

In 1994, there was a very interesting picture. Bill Gates was sitting on 330,000 sheets of paper with a compact disc, and proudly announced that one compact disc could carry the content recorded on the entire paper. The capacity unit of this CD is MB level. In 2010, Jobs held an iPhone, claiming to be able to hold tens of thousands of discs. The unit of this capacity is probably GB.

(The picture comes from the Internet)

When the 5G era comes, what are the characteristics of the entire view data we see? In summary, the first is video and picture data, which are widely generated on various terminal devices such as mobile phones, vehicle terminals, drones, and game consoles; the second feature is that the data level will reach ZB level per day; The three characteristics are scattered, and terminal equipment may generate video and picture data at any location; the fourth characteristic is that for the entire data, the value density of video and picture data is relatively low. All of the above characteristics will bring great challenges to the entire data digitization. How do we deal with such challenges? How to use cloud computing to more calmly welcome the arrival of the above scenes? View computing products are product solutions designed to undertake such business scenarios.

View computing-cloud PaaS service for service (view) terminal computing

Alibaba Cloud's view computing product is positioned to provide nearby connection, storage, and computing PaaS services for view terminals. This product combines the characteristics of Alibaba Cloud's edge computing nodes and public clouds, and carries out a cloud-based design for massive view terminal devices. Among them, the edge computing node provides a cloud computing node with city-level coverage, which can be closer to the device terminal. At the same time, view computing also combines the characteristics of public clouds, making big data analysis, data persistent storage, and application deployment easier.

In order to achieve the design effect of the product, Cloud made 1609cc0694a866 three-tier design for the view computing product architecture:

first layer of 1609cc0694a878 is for view-oriented data processing to build an edge computing node architecture.

The entire node including the physical resources of the basic layer, hardware selection, and basic network architecture have been specifically designed. For example, for large-scale view data analysis, the ability to cache data nearby, and the architecture to cache data (periodical storage) capabilities Special design optimization has been made to provide the ability to store nearby data with higher cost performance. Through periodic storage, first store the view data on the edge computing node for 1 day, 10 days or one month. When the data is to be used for further data applications, the data can also be synchronized to the public cloud region for further development. Data persistence. In the intermediate process, the high-density value data generated by AI inference calculations in the edge computing section can also be returned to the central region through the edge network for further big data processing.

At the same time, the node architecture also comes with some basic computing capabilities, including transcoding screenshots and so on.

second layer of end-side cloud collaborative layered computing architecture design

The so-called end-side cloud collaborative hierarchical computing is to effectively connect devices, edge computing nodes, and public clouds through the view computing PaaS platform, allowing data to be calculated at a suitable location.

In order to achieve this effect, the view computing has the connection capability of the device terminal in order to control the terminal. When some data needs to be processed in the terminal, the data can be processed in the terminal. At the same time, the nearest computing node can connect the data generated by the device terminal to the nearest connection for calculation. The data generated by the calculation can also be used for further large-scale applications and analysis through the public cloud.

In this way, the entire end-side cloud collaborative hierarchical calculation is formed, which can ensure the efficiency of the entire calculation.

The third layer is multi-node collaborative nearby computing.

Because the view data is generated at the terminal, customers prefer that the cloud node is close to the terminal to process the data, so the view computing needs to have the ability to coordinate multi-node computing to ensure that all terminal devices are connected to an edge computing node nearby, and Processing and analysis are performed at the edge computing node.

View computing builds the capabilities of location awareness and platform scheduling. The capabilities revealed by the platform can be location unaware, so that developers do not need to pay attention to the location of specific physical nodes, but only need to focus on the entire business process.

based on the above three-tier product architecture design, and view computing has three major product capabilities:

First, it has the ability to connect to device terminals.
You can use the one-click cloud open protocol platform that comes with View Computing to manage the terminal connection and data uplink; at the same time, you can also use standard methods, such as national standard, RTMP or Onvif, to connect the device terminal data to the nearest connection To the cloud. In order to better adapt to the developer’s flexibility, view computing has also made some custom design capabilities in the connection part, which means that developers can let data computing follow a custom protocol to connect the data of the device terminal to the cloud .

second, the nearby data storage capacity.
View computing provides the ability to cache nearby views, allowing periodic data to be stored nearby; at the same time, it also provides the ability to return data to the center, facilitating data persistence and large-scale data analysis.

third, AI computing power
View computing integrates the AI computing capabilities of Dharma Academy, and also comes with some basic computing capabilities, such as screenshots and transcoding. In order to make the processing of the entire video data more flexible and convenient, view computing also provides custom interfaces. Developers can integrate some algorithms developed by themselves or third parties into the view computing platform to achieve more flexible services. Development. At the same time, view computing also provides open and unified interfaces and open protocols to help developers achieve more flexible integration.

In order to achieve the above product capabilities, the underlying support includes technical modules such as scheduling, management, application hosting, monitoring, and operation and maintenance to achieve more efficient collaborative management, higher stability and reliability.

In the end, View Computing can with billions of terminal view data, allowing customers to use it out of the box and do simple and efficient development based on open interfaces.

Typical application scenarios of view computing

For various device terminals, customers of view-based computing products only need to do business processes, data application development, and final application deployment. The connection, storage and computing capabilities of other view data can all be undertaken through the view computing platform.

Speaking of specific scenarios, Alibaba Cloud has already implemented landing practices with partners in the following five scenarios:

first road computing scenario. based on the cloud-based practice of highway view data. The final result is that the entire vehicle traffic becomes safer and more monitorable. The use of view computing to build a smart high-speed system can allow traffic videos to be processed on the cloud nearby, and the video delay is reduced to 10 milliseconds. The AI calculation response time of traffic incident videos is increased by 70%, and the efficiency of highway safety incident processing is increased by 80%. Global traffic situation awareness, service level has been greatly improved;

second one of 1609cc0694aaff is the education scene, typical cloud classroom, which turns offline classrooms into digital classrooms. Through view computing products, remote online lectures and online learning can be realized; at the same time, view computing can also be done on view data. Further AI analysis and application, to help digital classrooms, digital classrooms view data, realize AI computing capabilities, improve students' class efficiency, help teachers better monitor education effects, and gradually improve their own teaching effects through feedback;

third is the new retail scene. the purchase link and supply chain link of the retail scene through view data, helping the retail scene improve the efficiency of sales and supply chain. Of course, there is a typical scenario in the supply chain that is logistics. View computing can help the logistics scenario. A full-link monitoring is done from the equipment, from the delivery of goods to the express delivery process, which improves management efficiency and allows consumers to enjoy more convenient shopping. Environment to help the digital upgrade of the new retail industry;

fourth one is public safety, includes public safety, food safety, bright kitchens and other video scenes; video-based localization of cloud and AI capabilities overlay, improve management efficiency, while reducing storage costs;

fifth home entertainment scene. can further enhance the effect of entertainment through the empowerment of view data, such as cloud games and AR/VR.

View-calculated product value

First, view computing is based on edge computing nodes, with nearby coverage and processing capabilities. has lower overall network costs, higher access flexibility, and can achieve low latency.

Second, the data processing of view calculation is hierarchical. First, the data can be roughly calculated on the terminal; secondly, the data can be calculated on the edge computing node, and further structured to extract higher-density value data; again, these data can also be transmitted back to the public cloud to make it bigger Large-scale data application and analysis.

At the same time, the data storage of view calculation also supports tiering. On the device terminal, the data can be temporarily stored; when the business needs to be stored periodically, the edge computing node can be selected for further periodic storage of the data; and when part of the data is more important, persistent data storage is required , You can store the data in the public cloud for further persistent storage through the scheduling ability of the data computing platform.

The benefits it brings are lower costs, as well as an improvement in the flexibility and convenience of expansion.

Third, another feature of view computing is the location-insensitive PaaS layer cloud service. are uniformly revealed to customers and developers through the view computing product, and behind the product are connected massive devices of view terminals, connected edge computing nodes, and connected to the public cloud. Therefore, the processing location of the data can be insensitive, and the appropriate node can do the most appropriate and most efficient computing service.

The benefits of such product design are lower network costs and higher development efficiency.

Fourth, view computing not only provides a one-click open protocol to the cloud, to help the entire device terminal to make convenient connection and access, but also open a programmable interface to help developers to access the protocol Make your own definition. For the entire AI computing capabilities, you can integrate the AI capabilities of Alibaba Cloud Dharma Academy, or you can implement AI computing in more subdivided scenarios through custom methods. The result is more flexible business development, higher development efficiency, and lower cost.

Future outlook

At the end of the online sharing, Yunjue said: I hope that more ecological partners (node layer, algorithm layer) will develop and implement with Alibaba Cloud, so that more business scenarios can use the convenience brought by view computing. . At the same time, it is also expected that view computing has smarter scheduling and smarter management capabilities, helping massive equipment terminals to achieve very convenient nearby access, storage and calculation, to tap the value of broader view data.

Follow-up Alibaba Cloud View Computing will share more latest product capabilities, solutions and technical practices in the "Alibaba Cloud Edge Plus" public account. Welcome everyone to discuss together.

Copyright Statement: content of this article is contributed spontaneously by Alibaba Cloud real-name registered users. The copyright belongs to the original author. The Alibaba Cloud Developer Community does not own its copyright and does not assume corresponding legal responsibilities. For specific rules, please refer to the "Alibaba Cloud Developer Community User Service Agreement" and the "Alibaba Cloud Developer Community Intellectual Property Protection Guidelines". If you find suspected plagiarism in this community, fill in the infringement complaint form to report it. Once verified, the community will immediately delete the suspected infringing content.