Practice of Jinshan Cloud real-time computing platform based on Apache Flink and its application in epidemic prevention scenarios


Chinese Session 2022-07-29 16:50 GMT+8  #streaming

Kingsoft Cloud real-time computing platform was established at the beginning of 2015, aiming to build a one-stop real-time computing development platform, reduce the threshold for development and operation personnel to directly use the open source engine for learning, operation and maintenance, and improve development efficiency. The platform base is based on Apache Flink, providing the ability to develop, test, deploy, and operate real-time computing applications in one stop. At present, in the financial, medical, Internet and other industries to serve a large number of enterprise customers. This sharing will be divided into two parts. The first part introduces the practice and thinking of Kingsoft cloud real-time computing platform in terms of product and architecture. The second part will introduce how to use the ability of Apache Flink computing engine to improve the timeliness of all kinds of epidemic prevention data under the background of special big data construction of epidemic prevention in a city, and quickly respond to the real-time computing needs of all kinds of epidemic prevention data relying on the real-time computing platform of Kingsoft Cloud. It provides more real-time data support for the large screen display of real-time data of epidemic prevention departments and various epidemic prevention work.

Share the outline: A. Development history of Kingsoft Cloud real-time computing platform B. Real-time computing platform product architecture and core functions C. Technical architecture of real-time computing platform D. Introduction to the realization principle of core functions E. Application in epidemic prevention scenarios F. Future development

Share key points: This paper introduces the product form and technical architecture of real-time computing platform. The core functions of the platform such as Flink On K8S and online fault diagnosis are deeply introduced. This paper introduces the demand for real-time computing in the context of epidemic prevention, and how to use ApacheFlink to quickly realize these demands and provide stable real-time computing data export on Kingsoft Cloud real-time computing platform.


Shuli Zheng: Kingsoft cloud group, Research and development experts, 10 + years of distributed system and large data real-time computing platform for business and development experience, in 2015 to join the kingsoft cloud group, has been engaged in real-time computing related work, now the kingsoft cloud group head of real-time computing platform, large data department is mainly responsible for kingsoft cloud group of real-time computing platform and data acquisition platform architecture and product design, research and development of the core modules and problems solving; Good at complex distributed system performance tuning and usability improvement; Focus on Flink real-time computing technology and big data platform architecture, willing to promote and preach real-time computing related technologies. DTCC (China Database Technology Conference) 2021 Lecturer