You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "Márton Balassi (JIRA)" <ji...@apache.org> on 2014/10/27 12:08:33 UTC

[jira] [Created] (FLINK-1190) Enhanced windowing semantics

Márton Balassi created FLINK-1190:
-------------------------------------

             Summary: Enhanced windowing semantics
                 Key: FLINK-1190
                 URL: https://issues.apache.org/jira/browse/FLINK-1190
             Project: Flink
          Issue Type: Improvement
          Components: Streaming
    Affects Versions: 0.8-incubating
            Reporter: Márton Balassi


The current windowing, batching options are rather strict: the stream is split into smaller units based on either count or time. This model is not flexible enough to e.g. describe the following semantics: provide the data from the last one second, but provide at most 1000 records.

The concept is based on IBM InfoSphere Stream's eviction and triggering policies, while giving the user the opportunity to define policies.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)