Re: A proposal about skew data handling in Flink

2015-10-20 Thread Fabian Hueske
ssue, and decide whether it's deserved to be resolved at the cost. > > Thanks > Chengxiang > -Original Message- > From: Fabian Hueske [mailto:fhue...@gmail.com] > Sent: Monday, October 19, 2015 7:15 PM > To: dev@flink.apache.org > Subject: Re: A proposal about ske

RE: A proposal about skew data handling in Flink

2015-10-19 Thread Li, Chengxiang
same time, Flink runtime assume that all tasks >> consume same size resources, this's not always true. Skew data >> handling try to make skewed data fit into Flink's runtime. >> I write a proposal about skew data handling in Flink, you can read it >> at >>

Re: A proposal about skew data handling in Flink

2015-10-19 Thread Fabian Hueske
consume same size resources, this's not >> always true. Skew data handling try to make skewed data fit into Flink's >> runtime. >> I write a proposal about skew data handling in Flink, you can read it at >> https://docs.google.com/document/d/1ma060BUlhXDqeFmviEO7Io

Re: A proposal about skew data handling in Flink

2015-10-16 Thread Fabian Hueske
t; Flink runtime assume that all tasks consume same size resources, this's not > always true. Skew data handling try to make skewed data fit into Flink's > runtime. > I write a proposal about skew data handling in Flink, you can read it at > https://docs.google.com/document/d/1

A proposal about skew data handling in Flink

2015-10-15 Thread Li, Chengxiang
ile at the same time, Flink runtime assume that all tasks consume same size resources, this's not always true. Skew data handling try to make skewed data fit into Flink's runtime. I write a proposal about skew data handling in Flink, you can read it at https://docs.g