yihua opened a new issue, #20251: URL: https://github.com/apache/hudi/issues/20251
Bulk insert through the row writer (the default for bulk_insert) ignores `hoodie.parquet.bloom.filter.enabled=false` and always writes the parquet footer bloom filter. `HoodieInternalRowFileWriterFactory` checks only whether meta fields are populated, while the Avro, Spark and Flink writer factories use the shared `HoodieFileWriterFactory.enableBloomFilter` added with the config in #10457. With default settings the bloom adds about 430 KB to every base file footer, which every reader loads and parses. Proposal: have the row-writer factory call the shared `enableBloomFilter` / `createBloomFilter`. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
