Aegeaner的专栏

Flink JOIN 执行计划代码： val table1 = env.fromElements((1, "hello")).toTable(tEnv, 'a, 'b) val table2 = env.fromElements((1, "hello")).toTable(tEnv, 'c, 'd) val table = table1.join(table2).where("

摘自Flink-forward的一张幻灯片，也是衡量流计算平台的四个很好的基准和标尺。

背景：Flink 1.2起，对流处理中耗时较高的外部IO操作进行了优化，引入了Async IO特性，将IO操作异步化，大幅提升了效率。测试环境：Java HotSpot(TM) 64-Bit Server VM 1.8.0_112-b16 on Mac OS X 10.11.6Intel(R) Core(TM) i5-5257U CPU @ 2.70GHz 测试基准：对

https://fpj.me/2017/07/04/no-consensus-in-exactly-once/JULY 4, 2017 ~ FPJExactly-once semantics is an intriguing, controversial, and for me an exciting topic. I have been dealing with it

13 Mar 2015 by Fabian Hüske (@fhueske)Join Processing in Apache FlinkJoins are prevalent operations in many data processing applications. Most data processing systems feature APIs that make jo

By Stefan Richter (@StefanRRichter) of data ArtisansNow, it’s time for a deep-dive on application state in Flink. We’ll cover 5 topics in this section:An Intro to Stateful Stream Processin