Hi Devs,
Thank you all for the valuable feedbacks and comments in the previous
design doc.
We have currently created an initial design/work plan based on @Kurt's
suggestions as "improvements with the sliding window scenario without
changing/adding new public APIs".
Please kindly take a look at
Hi Kurt,
Thanks for the valuable feedback. I think the suggestions you and Jincheng
provided are definitely the best execution plan
- Starting with sliding window optimization by exhausting the current
public API, there are some components we can leverage or directly reuse
from Blink's window
Hi Rong,
Thanks for the detailed summarization! It indeed involves lots of problems
and unanswered questions which is i think not practical to
solve in one shot. From my point of view, the performance issue with the
sliding window is the root one and maybe most possible
which user will run into.
Hi All,
Thanks for sharing feedbacks for the window optimization design doc and on
the discussion JIRAs @Jincheng, @Kurt, @Jark and @Fabian. These are very
valuable feedbacks and we will try to incorporate them in the next step.
There were several revision done for the current design doc, and
Hi Rong,
Thanks for the improvement proposal, this topic aroused my interest since
we did some similar improvements in Blink.
After going through your design doc, i would like share some thoughts:
1. It looks to me the proposed SliceManager and MergeManager is quite
generic and can be used in
Thank you Rong!
The performance of sliding windows is an issue for many users.
Adding support for a more efficient window is a great effort.
Thank you,
Fabian
Am Di., 29. Jan. 2019 um 16:37 Uhr schrieb Rong Rong :
> Hi all,
>
> Thanks for the feedbacks and suggestions to the design doc. I have
Hi all,
Thanks for the feedbacks and suggestions to the design doc. I have created
a parent JIRA [1] to track the related tasks and started the implementation
process
Any further feedbacks or suggestions are highly appreciated.
Best,
Rong
[1] https://issues.apache.org/jira/browse/FLINK-11276
Hi all,
Various discussion in the mailing list & JIRA tickets [2] had been brought
up in the past regarding the windowing operation performance. As we
experiment internally with some of our extreme use cases, we found out that
using a slice-based implementation can optimize Flink's windowing