[
https://issues.apache.org/jira/browse/FLINK-22136?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Arvid Heise updated FLINK-22136:
--------------------------------
Description:
To test unaligned checkpoints, we should use a few different applications that
use different features:
* Mixing forward/rescale channels with keyby or other shuffle operations
* Unions
* 2 or n-ary operators
* Associated state ((keyed) process function)
* Correctness verifications
The sinks should not be mocked but rather should be able to induce a fair
amount of backpressure into the system. Quite possibly, it would be a good idea
to have a way to add more backpressure to the sink by running the respective
system on the cluster and be able to add/remove parallel instances.
Things to check in the application
* Inflight data is restored to the correct keygroups -> can be checked with
keyed state in a process function
* Correctness: Completeness (no lost records) + no duplicates
* Orderness of data for keyed exchanges (we guarantee that records with the
same key retain orderness across keyed operators)
* (To detect errors early, we can also use magic headers)
was:
To test unaligned checkpoints, we should use a few different applications that
use different features:
* Mixing forward/rescale channels with keyby or other shuffle operations
* Unions
* 2 or n-ary operators
* Associated state ((keyed) process function)
* Correctness verifications
The sinks should not be mocked but rather should be able to induce a fair
amount of backpressure into the system. Quite possibly, it would be a good idea
to have a way to add more backpressure to the sink by running the respective
system on the cluster and be able to add/remove parallel instances.
> Device application for unaligned checkpoint test on cluster
> -----------------------------------------------------------
>
> Key: FLINK-22136
> URL: https://issues.apache.org/jira/browse/FLINK-22136
> Project: Flink
> Issue Type: Sub-task
> Reporter: Arvid Heise
> Priority: Major
>
> To test unaligned checkpoints, we should use a few different applications
> that use different features:
> * Mixing forward/rescale channels with keyby or other shuffle operations
> * Unions
> * 2 or n-ary operators
> * Associated state ((keyed) process function)
> * Correctness verifications
> The sinks should not be mocked but rather should be able to induce a fair
> amount of backpressure into the system. Quite possibly, it would be a good
> idea to have a way to add more backpressure to the sink by running the
> respective system on the cluster and be able to add/remove parallel instances.
> Things to check in the application
> * Inflight data is restored to the correct keygroups -> can be checked with
> keyed state in a process function
> * Correctness: Completeness (no lost records) + no duplicates
> * Orderness of data for keyed exchanges (we guarantee that records with the
> same key retain orderness across keyed operators)
> * (To detect errors early, we can also use magic headers)
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
