Offset overflow - Native Parquet scan expands flat dictionary columns and reads twice Spark's batch rows

Ouverte
  • Quelque chose ne va pas ?
#7,499 0 commentaires 0 réactions 0 personnes assignées Voir sur GitHub

Les mainteneurs répondent en général sous 2 jours

Personne n'a encore pris cette issue.

Évaluation

Difficulté
5/5
Temps estimé
Plus d'une semaine
Accessibilité débutants
42/100
Type d'issue
Bug
Clarté
Plutôt claire
Activité
Active
Stack technique
rust, scala, spark
Domaine
data-engineering, performance

Piste de recherche

Start with the native Parquet scan and the arrow-rs byte-array decoding paths described in the issue, including the DELTA_LENGTH_BYTE_ARRAY decoder. Reproduce the overflow with a flat dictionary-encoded string column and a small injected byte budget, then compare native scan results with Spark's reader. Done means byte-bounded batching or dictionary handling prevents overflow while preserving the expected rows; the issue also calls out matching Spark's default batch size and an upstream decoder panic fix.

Rédigé par le modèle d'indexation à partir du texte de l'issue.

Description

area:scan bug priority:medium

Part of #7267.

Problem

Two things make the native Parquet scan, and the native Iceberg scan that uses the same arrow-rs reader, overflow a string or binary column where Spark's reader does not.

Flat dictionary-encoded columns are expanded per batch. Comet asks arrow-rs for plain Utf8/Binary, so the reader copies a dictionary value into the batch once per row that references it (extend_from_dictionary, parquet 59.3.2 src/arrow/buffer/offset_buffer.rs:92-118). A column with a few wide distinct values is stored as a small dictionary page plus tiny keys, and can still expand past 2 GiB in one 8192-row batch: one 300 KB value repeated 8192 times is 2.4 GB. The scan fails with index overflow decoding byte array. Spark's vectorized reader keeps the dictionary ids and decodes a value only when a row reads it, so it succeeds. #7254 covers the nested case; flat columns overflow inside arrow-rs instead.

Twice the rows per batch. Comet reads spark.comet.batchSize rows per batch, 8192 by default. Spark's vectorized Parquet reader reads spark.sql.parquet.columnarReaderBatchSize rows (4096), and Iceberg's Java reader 5000. With values of 256-512 KiB in row groups large enough to fill a batch (large row-group sizes or highly compressible values), Comet's batch passes 2 GiB and Spark's does not.

The DELTA_LENGTH_BYTE_ARRAY decoder panics instead of returning that error (parquet 59.3.2 src/arrow/array_reader/byte_array.rs:514).

The epic lists the scan as honoring the batch size. It does, but in both cases the batch size alone doesn't bound the bytes.

Workaround

Lower spark.comet.batchSize.

Proposed fix

  • Keep dictionary-encoded byte columns as dictionaries out of the reader and unpack them after the scan in row ranges bounded by bytes, or cap scan batches by bytes.
  • At least, default the native scan's batch size to Spark's reader batch size.
  • Upstream: return an error from the DELTA_LENGTH_BYTE_ARRAY decoder instead of panicking.

Tests

A Parquet file with a flat dictionary-encoded string column, one 300 KB value repeated 8192 times, stays tiny on disk. Reading it with the native scan and comparing with Spark shows the problem, but needs about 2.5 GB of memory, so test the byte-bounded batching with a small budget injected for the test.

Langage dominant
Scala
Étoiles
1.3k
Forks
414
Merge moyen
3 j 4 h
PR mergées (30 j)
202

Préparer son environnement

Par où commencer

  1. Lisez l'issue en entier, puis le guide de contribution du projet.
  2. Signalez en commentaire que vous la prenez — cela évite que deux personnes fassent le même travail.
  3. Forkez le dépôt et travaillez sur une branche.
  4. Ouvrez une pull request qui référence le numéro de l'issue.

Autres issues de apache/datafusion-comet

Toutes les issues de apache/datafusion-comet

Issues similaires

Plus d'issues Scala

Recevez les nouvelles issues par e-mail

Un résumé court des issues GitHub adaptées aux débutants.