A Year of Blink at Alibaba: Apache Flink® in Large Scale Production
Earlier this month, Dataversity published a follow-up post from Xiaowei Jiang, a Senior Director on Alibaba’s search infrastructure team, where he gave an update on his team’s progress both with Blink (Alibaba’s Apache Flink-based system) and with Alibaba’s open source contributions to Flink.
The Dataversity post picks up where an earlier post from Xiaowei leaves off. In October 2016, Xiaowei published a blog post about Blink on the data Artisans blog where he introduced how Blink powers various components of Alibaba’s search infrastructure and why Alibaba chose a Flink-based system. He also detailed some of the enhancements that his team had added to Blink so that it was a better fit their use case.
The two posts provide an inside view of how Blink has moved into an increasingly central role at Alibaba in the past year while at the same time, the Alibaba team has become an active contributor to the open source Flink community.
And if you’re interested in learning more, we recommend Alibaba’s talks from Flink Forward San Francisco 2017:
- Runtime Improvements in Blink for Large-scale Streaming at Alibaba
- Blink’s Improvements to Flink SQL And TableAPI