Hi,
In my flink operators, I need to connect to an external service to update
state. I was thinking that the updates to the external service can be
synchronized via checkpoint barriers.

The topology of the stream is a source, then a single stage of operator
replicas handling different partitions, then all joining in a single sink.

Each operator will contact the external service when it receives a
checkpoint barrier and uploads local state (which caches the uploads and
returns a handle).

After upload, it forwards the cache handle to the sink. Once sink receives
handles from all such operators, it calls the external service with a list
of handles received. This helps ensure that all handles are from the same
checkpoint barrier.

Is it possible to achieve this in a flink application?

Thanks,
Gopi

Reply via email to