Hello Team,
I have a scenario in Elasticsearch transform where I need to ignore all duplicate items from the source index and apply few aggregation after ignore the duplicate documents.
Could you please suggest any best design for this issue??
Thank you in advance
Regards
Venkatesh