You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hudi.apache.org by "Rajesh Mahindra (Jira)" <ji...@apache.org> on 2025/07/23 17:44:00 UTC

[jira] [Created] (HUDI-9627) Add bloom filter pruning when looking up keys in metadata files

Rajesh Mahindra created HUDI-9627:
-------------------------------------

             Summary: Add bloom filter pruning when looking up keys in metadata files
                 Key: HUDI-9627
                 URL: https://issues.apache.org/jira/browse/HUDI-9627
             Project: Apache Hudi
          Issue Type: Improvement
            Reporter: Rajesh Mahindra


When performing key look up in metadata (hfile) files, we only leverage the hfile pruning (based on ranges). However, we are already writing bloom filters for each hfile. This PR adds an optional ability to first filter the keys based on the bloom filters, which can avoid the download and deser of a subset of the Hfile blocks and reduce latencies. This should be essentially useful for upsert workloads with mostly new(insert) records.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)