Review | Apache Flink meetup Hangzhou station ends successfully (PPT download attached)

Time:2020-7-24

On May 16, 2020, the first online live broadcast of Apache Flink meetup · Hangzhou station was successfully completed. This meetup invited four technical experts from kangaroo cloud, Netease cloud music, youzan and Alibaba to share their views on real-time data warehouse, 1.10 production environment practice, Flink distributed synchronization tool and Flink’s application in AI process.

<u><p style=”text- align:center “> meetup live broadcast review: < / P > < / u >
<p style=”text-align:center”>https://developer.aliyun.com/…</p>
<p style=”text-align:center”><u> Ppt Collection Download</p>
<p style=”text-align:center”>https://developer.aliyun.com/…</p>

Flinkx: a distributed data synchronization tool for kangaroo cloud based on Flink

Xu Chao Flink contributor, senior technical expert of kangaroo cloud

Review | Apache Flink meetup Hangzhou station ends successfully (PPT download attached)

Guest introduction:

Xu Chao, senior technical expert of kangaroo cloud and Flink contributor, is responsible for kangaroo cloud’s original one-stop data center PAAS – product research and development of datastack from 0 to 1; participated in the application of digital stack products to large-scale government and enterprise, finance, education, new retail and other industries, and was responsible for the technical services of typical customers such as BOE, China Merchants Securities, Ningbo Development and Reform Commission, Zhejiang University, and Piao pay. Currently, it focuses on the peripheral expansion of Flink real-time computing, including kangaroo cloud flinkx and Flink stream SQL Open source projects.

Introduction to the speech:

1、 Why Flink based implementation

  1. Distributed, simple concurrent configuration
  2. Easy to expand
  3. Efficiency

2、 Overall structure (including architecture diagram)
3、 How to control concurrency and limit current (principle and Implementation)
4、 Incremental synchronization, breakpoint continuous transmission (principle and Implementation)
5、 Real time acquisition (principle and Implementation)
6、 Current plug-in support and future development plan

“Netease cloud music real time data warehouse practice”

Wang Lei, Netease data platform development expert

Review | Apache Flink meetup Hangzhou station ends successfully (PPT download attached)

Guest introduction:

Wang Lei, an expert in the development of Netease data platform, is currently responsible for the construction of cloud music real-time, offline and machine learning platforms.

Introduction to the speech:

With the development of cloud music business, the demand for real-time computing business is more and more high, and the platform is facing more and more challenges. In order to reduce the threshold of real-time computing development, better control of real-time tasks, and improve the stability of the system, we integrated our existing underlying services and developed a real-time data warehouse solution suitable for cloud music.

This sharing mainly introduces our overall practice plan, some problems we are facing, and some future plans. The overall outline is as follows:

  1. Introduction to cloud music overall data platform architecture
  2. Cloud music Flink real time data warehouse and metadata center integration
  3. Current problems and challenges
  4. Practice and planning of batch flow integration

Practical application of Apache Flink SQL 1.10 in youzan

Shen Lei, R & D Engineer of youzan data

Review | Apache Flink meetup Hangzhou station ends successfully (PPT download attached)

Guest introduction:

Shen Lei, whose English name is lakeshen, is currently in charge of the development and maintenance of the real-time computing platform and Flink computing engine in youzan. I actively participate in the Flink open source community and like to share Flink technology with you. I have written a lot of Flink related technical articles to help you learn Flink together.

Introduction to the speech:

This speech mainly explains the practical application of Flink SQL version 1.10 and why Flink SQL is chosen. The main contents are as follows:

  1. Real time platform Flink SQL development practice
  2. Practice and optimization of Flink connector
  3. Flink SQL has great business support

Application of Apache Flink in AI process

Tu Yiwen (reply) | Alibaba Development Engineer

Review | Apache Flink meetup Hangzhou station ends successfully (PPT download attached)

Guest introduction:

Tu Yiwen, Huaming Yingyan, joined Alibaba in July 2019 as a development engineer, focusing on the development of Flink AI ecological field yiwen.png 。

Introduction to the speech:

In today’s AI scenarios, Flink’s unique advantage in streaming mode brings more possibilities for AI development. In order to better support AI development in different scenarios, Flink community and Flink’s ecological projects are also making continuous efforts. This paper will introduce the application of Flink in AI scenarios, including:

  1. Background on Flink embracing AI
  2. What is AI flow
  3. Meet Flink AI flow
  4. An ML scenario

<p style=”text- align:center “> < U > group photo on site</p>

Review | Apache Flink meetup Hangzhou station ends successfully (PPT download attached)