Hi Biplob, given these requirements, I would rather not use a window but implement the functionality with a stateful ProcessFunction. A ProcessFunction can register timers, e.g., to remove inactive state. The state of a ProcessFunction can be made queryable.
Best, Fabian 2017-07-31 9:52 GMT+02:00 Biplob Biswas <revolutioni...@gmail.com>: > Thanks Fabian for the reply, I was reconsidering my design and the > requirement and what I mentioned already is partially confusing. > > I realized that using a sessionwindow is better in this scenario where I > want a value to be updated per key and the session resets to wait for the > gap period with every update after which it should be removed. > > From whatever I read currently about session windows (and windows in > general), is that I can perform aggregates over it. What I was thinking was > that rather than performing an aggregate or reduce, I can simply replace > the > old value. > > now my problem is, would such a state be queryable? > > As you mentioned that siding windows are defined on streams, is it the same > for session windows? > > "I basically want a state which updates by key, and where records can > self-destruct after a fixed amount of time(basically records being part of > a > session) if not updated (such that the state doesn't grow indefinitely)?" > > Are there any abstractions to do the same and if not, is it possible to > discuss if this can be done otherwise? > > Thanks a lot, > Biplob > > > > -- > View this message in context: http://apache-flink-user- > mailing-list-archive.2336050.n4.nabble.com/Flink- > QueryableState-with-Sliding-Window-on-RocksDB-tp14514p14549.html > Sent from the Apache Flink User Mailing List archive. mailing list archive > at Nabble.com. >
