[SPARK-56803][SQL] Add bulk read+narrow path for INT64 DECIMAL to 32-bit Decimal Parquet vector updater#55853
Draft
LuciferYang wants to merge 2 commits into
Draft
[SPARK-56803][SQL] Add bulk read+narrow path for INT64 DECIMAL to 32-bit Decimal Parquet vector updater#55853LuciferYang wants to merge 2 commits into
LuciferYang wants to merge 2 commits into
Conversation
…bit Decimal Parquet vector updater
LuciferYang
force-pushed
the
SPARK-56803-downcast-long
branch
from
800ed85 to
39b038d
Compare
…uet.ParquetVectorUpdaterBenchmark (JDK 17, Scala 2.13, split 1 of 1)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
Extend the bulk read+widen pattern introduced in SPARK-56791 to
DowncastLongUpdater(parquet INT64 + DECIMAL(p<=9) read into a Spark 32-bitDecimalType).A new
readLongsAsIntsdefault method onVectorizedValuesReaderdoes the per-row fallback.VectorizedPlainValuesReaderoverrides it to fetch source bytes once viagetBuffer(total * 8)and run a tight in-method conversion loop.DowncastLongUpdater.readValuesbecomes a one-line delegation. The narrowing is Java's primitive long-to-int cast ((int) buffer.getLong()), which discards the high 32 bits; this is non-lossy in practice because Parquet's DECIMAL(p<=9) encoding bounds the value range to[-999_999_999, 999_999_999].Why are the changes needed?
DowncastLongUpdater.readValuesallocates a freshByteBufferslice insidegetBuffer(8)for every element on the legacy path, and that allocation dominates the loop. Collapsing N allocations into one is the same win SPARK-56791 delivered for the INT32 -> Long sibling.Does this PR introduce any user-facing change?
No.
How was this patch tested?
(To be updated after the GHA benchmark and test runs complete.)
Was this patch authored or co-authored using generative AI tooling?
Generated-by: Claude Code