[ https://issues.apache.org/jira/browse/FLINK-7001?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16464544#comment-16464544 ]
Shuyi Chen commented on FLINK-7001: ----------------------------------- Hi [~pgrulich], the paper is a nice read. And the technique applies to Tumble, Sliding & Session window, which is a good win, and the evaluation result looks good. Also, it seems you already have an implementation for Scotty using Apache Flink based on the paper. Maybe, you and [~jark] can share more, for each approach, about the detail design, pros and cons, and we can discuss them here? > Improve performance of Sliding Time Window with pane optimization > ----------------------------------------------------------------- > > Key: FLINK-7001 > URL: https://issues.apache.org/jira/browse/FLINK-7001 > Project: Flink > Issue Type: Improvement > Components: DataStream API > Reporter: Jark Wu > Assignee: Jark Wu > Priority: Major > > Currently, the implementation of time-based sliding windows treats each > window individually and replicates records to each window. For a window of 10 > minute size that slides by 1 second the data is replicated 600 fold (10 > minutes / 1 second). We can optimize sliding window by divide windows into > panes (aligned with slide), so that we can avoid record duplication and > leverage the checkpoint. > I will attach a more detail design doc to the issue. > The following issues are similar to this issue: FLINK-5387, FLINK-6990 -- This message was sent by Atlassian JIRA (v7.6.3#76005)