IndexTables

May 16, 2026 min read

IndexTables is an open table format for Apache Spark that enables fast retrieval and full-text search across large-scale data. It integrates seamlessly with Spark SQL, combining powerful search capabilities with joins, aggregations, and standard SQL operations.

Contribution

docs: replace deprecated mergeGroupMultiplier in features.md

PR #335Merged May 2026

Replaces the deprecated spark.indextables.mergeOnWrite.mergeGroupMultiplier configuration key with the current spark.indextables.mergeOnWrite.minBatchesToTrigger in the features reference documentation, closing issue #237.

The deprecation note in configuration.md directs users from mergeGroupMultiplier to the batchCpuFraction + minBatchesToTrigger pair. minBatchesToTrigger=2 is the closest 1:1 substitute for mergeGroupMultiplier=2.0 — both raise the merge trigger threshold by 2× over the default — making this a clean, low-ambiguity migration for existing users.