Support Dedup of objects in Stream Output #11430
Labels
distributed framework
enhancement
Enhancement or improvement to existing feature or request
Indexing
Indexing, Bulk Indexing and anything related to indexing
Performance
This is for any performance related enhancements or bugs
Is your feature request related to a problem? Please describe.
Today, if we write multiple primitives/objects to our streams, we've observed that the objects get duplicated always. We can save a lot of network and serialization/deserialization CPU if we can de-dup based on the references using a.
Describe the solution you'd like
Something similar to HandleTable in ObjectOutputStream used in JDK
Describe alternatives you've considered
Write custom logic for data structures and objects but that may not cover all the use cases.
Additional context
The text was updated successfully, but these errors were encountered: