Open source: the recipe for success in Big Data systems
- LecturerDr. Wei-Chiu Chuang (Tech Lead of the big data storage system, HDFS, at Cloudera. Hadoop, (USA))
Host: Chien-Min Wang - Time2019-07-09 (Tue.) 10:00 ~ 12:00
- LocationAuditorium106 at IIS new Building
Abstract
In a world filled with AI hypes, it is easy to get lost and be oblivious to the fact that Computer Science fundamentals are still as important as ever before, if not more.
Dr. Chuang is the Tech Lead of the big data storage system, HDFS, at Cloudera. Hadoop, the de-facto standard for Big Data processing, is essential for providing the computer and storage power to the modern machine learning platforms. My job is to ensure Hadoop can scale up to thousands of nodes, hundreds of petabytes of data for hundreds of applications. A solid Computer Science background is critical in this job because we see hard problems every single day.
Making sure systems run reliably at scale and performant is a challenging task. The Open Source development model is the receipt for success because we are able to freely work with the top distributed systems talents around the world, and then come up with the best solutions to solve them. It is also a rewarding experience to learn from these great engineers.
In this talk, He will share my personal story in open source collaboration, and how he applied systems research into Big Data systems development.