Time: 2019.06.29 (Sat) 10: 00-18: 00
Address: Beijing Chaoyang Wangjing Posco Center building b seat two-story multi-purpose hall W1
Registration link: https://www.huodongxing.com/event/9492578872100
Guest Information
Session I: "Apache Flink 1.9 Characteristics of interpretation"
Dai financial viability (Gordon Tai)
Apache Flink PMC, Ververica Software Engineer
Profile : Dai-funded force (Gordon Tai) is a member of PMC Apache Flink open source community, is currently working in Ververica as Software Engineer, is responsible for system development Flink's. Flink major contribution comprises: Apache Kafka / AWS Kinesis a precise connection data sources, data serialization framework class, have application upgrade state of stream processing. Had Flink Forward San Francisco / Berlin / Beijing and Strata Data as speakers to share Flink related issues.
Yang Specter (nickname: Rooney)
the Apache Flink the Committer's, Alibaba senior technical experts
biography : in 2011 graduated from the Zhejiang University to join Alibaba, we have engaged in a search engine, scheduling systems, large data processing core of the system design and research and development, currently responsible for real-time computing Flink SQL engine.
Session II: "Flink Table API to create based on machine learning ecology"
Qin Jiangjie
Apache Kafka PMC, Alibaba senior technical experts
Profile : Alibaba real-time computing platform for high-level technical experts. Graduated from the Carnegie Mellon University, worked on the development of LinkedIn responsible Apache Kafka, it is the Apache Kafka PMC member. Alibaba is currently involved in the development of Apache Flink.
Description : Flink Table API community to a series of enhanced features in the most recent period of time to make it better for a variety of computing needs of data services. Machine learning as an important computing scenarios also Table API key component of development planning. The share Flink will introduce the community to build a machine learning progress of the work and ecological planning based Flink Table API.
Session III: "Flink on Kubernetes based big data platform"
Zhang Kai
Ali cloud Advanced Technical Expert
Profile: Responsible for container services Kubernetes product architecture and development, focusing on exploring the use of container technology to accelerate heterogeneous computing, deep learning, edge computing program delivery and a wide range of scenes landing.
Renchun De (nickname: WALL)
Alibaba senior technologist
Profile : graduated in 2006 joined Alibaba Group, has long been engaged Hadoop big data related to computing platform research and development, currently serves as senior technical experts in computing platforms division, responsible for YARN resources and Flink scheduling research and development. YARN by Flink and the depth of docking, to provide real-time calculation of large-scale, efficient and stable operating platform.
Session IV: "Based on high-performance machine learning algorithm library Apache Flink"
Yang
Alibaba algorithm senior expert
Profile : Yang is calculated Alibaba Group senior expert algorithm Platforms Group, is responsible for the development and maintenance of basic machine learning algorithms of machine learning algorithms Ali cloud platform in the PAI, based on research and development and Flink batch flow one common algorithm platform Alink. Alink has been widely used within the Ali Group, Yang and his team recently Flink is promoting open source-based machine learning algorithm library, further back to the community.
Speaker Profile: Alibaba computing platform Flink Division is working with the community, open from research in machine learning algorithm library, based on the algorithm library, users can build a high-performance machine learning Flink jobs more easily. We want to promote the development Flink community in the field of machine learning through open source. Also welcome more developers to join hands with us to build stronger, more complete Flink algorithm library. The share is mainly based around the team accumulated technology and harvest Flink development of high performance machine learning algorithm library process.
Five topics: "Apache Flink application and practice in the deft"
Dong Kingston
deft real-time calculation engine team leader
Profile : Dong Kingston, deft big data architecture real-time calculation engine team leader. Flink engine is currently responsible for research and development, application and construction of the peripheral subsystem in deft. 2013 graduated from the Dalian University of Technology, worked at odd tiger 360,58 Group. The main research areas include: distributed computing, scheduling system, distributed storage systems.
Speaker Profile : Flink system deft has a wide range of applications, including real-time monitoring, real-time statistical analysis, multi-source data join, involving almost all of the data related to the company's business units. Currently the overall size Flink cluster of over a thousand units daily processing entries over 26.4 billion, more than 360 ten million deal with peak / s. This share will introduce Flink system related problems in application practice deft, and we encountered stability, performance and Solutions. For example, how to solve performance bottlenecks rockdb backend interval join the next scene, a large number of stability issues when reading historical data speed difference between the number of multiple data sources to take the lead and so on. Specific outline:
1, Flink scenario and scale deft
2, optimization and improvement of engine deft Flink
3, future plans
Six topics: "Airbnb real-time streaming architecture and evolution of computing."
Wang Hao
Airbnb Data Platform Engineering Manager
Profile : Wang Hao, Ph.D., University of Southern California, now working Airbnb, the incumbent Data Platform Engineering Manager, led the stream computing team to provide real-time streaming computing infrastructure for the whole company. Worked at IBM Watson, Watson is responsible for building support for product data platform. The main areas for real-time stream computing, big data analysis systems, natural language processing.
Seven topics: "Apache Flink-1.9 Compatibility with Hive"
Li Rui
Apache Hive PMC, Alibaba technical experts
Profile : Ali Baba technical experts, Apache Hive PMC members, before joining Alibaba worked at Intel, IBM and other companies, mainly involved in Hive, HDFS, Spark and other open source projects.
Speaker Introduction : In order to improve Flink SQL functions to better tap the potential Flink in a batch regard, we propose Flink integration with Hive project, to provide users with Hive by Flink SQL interoperability features. This presentation will describe the design and architecture of the project, in Flink-1.9 integration with Hive in progress, and we plan to follow-up and so on.
Meeting Process
9: 00-10: 00: sign
10: 00-11: 00: Dai Yang funding of + Specter "Apache Flink 1.9 Characteristics of interpretation"
11: 00-11: 15: Coffee Break
11: 15-12: 00: Qinjiang Jie "to create a machine learning based Flink Table API's ecological"
13: 00-14: 00: sign
14: 00-14: 45: Zhang Kai + Renchun De "Flink on Kubernetes based big data platform"
14: 45-15: 30: Yang "based on high performance machine learning algorithm library Apache Flink"
15: 30-16: 15: Dong Ting-Ting "Apache Flink in the application and practice of deft"
16: 15-16: 35: Coffee Break
16: 35-17: 20: Wang Hao "Airbnb real-time streaming architecture and evolution of computing."
17: 20-18: 05: Rui "Apache Flink-1.9 and Hive Compatibility"
18: 05-18: 30: + Draw + group photo after the exchange
Scan code to join the community nails crowd, the first to get more first-hand information!
