Stefan Richter

icon indicating an email [email protected]

icon company Confluent icon location Berlin icon location Principal Engineer at Confluent. Ex-Snowflake, ex-Data-Artisans (original creators of Apache Flink), Ph.D. in Database Systems from Saarland University.

Results 4 comments of Stefan Richter

[FLINK-31238] [WIP] Use IngestDB to speed up Rocksdb rescaling recovery (rebased)

@rkhachatryan Can you take one more look? I have fixed all problems that I encountered in the original PR and also addressed your comments. Using `TestStreamEnvironment.randomizeConfiguration` seems a bit problematic...

[FLINK-34199] Add tracing for durations of rescaling/restoring RocksDB incremental checkpoints from downloaded and local state

I think the description is correct, depending on you point of view. Recovery today has two phases: 1. If there is any remote state it will be downloaded and becomes...

[FLINK-34199] Add tracing for durations of rescaling/restoring RocksDB incremental checkpoints from downloaded and local state

Besides that, I already merged a PR that measures local/remote state sizes. And in our meeting we found that all of the metrics are useful :)

[FLINK-34199] Add tracing for durations of rescaling/restoring RocksDB incremental checkpoints from downloaded and local state

I wrote the docs slightly different, hope that makes it clearer.