Page History

...

In addition to the above change, we will utilize be making use of the state backend optimization introduced in FLIP-325, where we . We will use a StateBackend state backend with a cache size of 1 if the cache size is not explicitly set by the users or use the cache size set by the users. The isn't explicitly specified by users so that we optimize the state backend access while minimizing the memory overhead. Alternatively, when users have set a cache size, we will adhere to that setting. This optimization can significantly enhance the performance of keyed one input operator should be improved considerably operators without code change when "isBacklog=true, such as KeyedProcessOperator, StreamGroupedReduceOperator, WindowOperator, etc. And ". Moreover, the performance of the operator during isBacklog=true is close to the performance in batch mode.

...

Currently, there are two keyed multi-input operators in DataStream API, KeyedCoProcessOperator and IntervalJoinOperator. Both of them will be optimized in the following way:

...