Issues · apache/parquet-java · GitHub

Labels Milestones New issue

Optimize dictionary writers by replacing fastutil Linked maps with OpenHashMap + ArrayList

#3513

· iemejia opened

on Apr 20, 2026

Add JMH benchmarks for encoding/decoding paths and fix parquet-benchmarks shaded jar

#3511

· iemejia opened

on Apr 19, 2026

Optimize BinaryPlainValuesReader by reading directly from ByteBuffer

#3509

· iemejia opened

on Apr 19, 2026

Reduce allocation overhead of no-op `GeospatialStatistics.Builder`

Type: enhancement

#3507

· arouel opened

on Apr 19, 2026

Optimize ByteStreamSplitValuesReader page transposition

#3505

· iemejia opened

on Apr 19, 2026

Optimize ByteStreamSplitValuesWriter: remove per-value allocation and batch single-byte writes

#3503

· iemejia opened

on Apr 19, 2026

Avoid per-write virtual dispatch in `DictionaryValuesWriter.shouldFallBack()` by caching the size-exceeded check

Type: enhancement

#3501

· arouel opened

on Apr 19, 2026

Cache hashCode() for non-reused Binary instances (huge dictionary-encode speedup)

#3499

· iemejia opened

on Apr 19, 2026

`DefaultValuesWriterFactory` uses thread-unsafe static singletons, causing cross-writer allocator contamination

#3497

· arouel opened

on Apr 19, 2026

Optimize PlainValuesWriter by writing directly to ByteBuffer slabs (up to 2x encode speedup)

#3495

· iemejia opened

on Apr 19, 2026

Optimize PlainValuesReader by reading directly from ByteBuffer (12x decode speedup)

#3493

· iemejia opened

on Apr 19, 2026

`Binary.copy()` does not materialize direct `ByteBuffer`-backed values to heap, causing use-after-free

#3489

· arouel opened

on Apr 18, 2026