This is an automated email from the ASF dual-hosted git repository.
MaxGekk pushed a commit to branch branch-4.x
in repository https://gitbox.apache.org/repos/asf/spark.git
The following commit(s) were added to refs/heads/branch-4.x by this push:
new 3d2b220ee59b [SPARK-56895][SQL] Batch ByteBuffer slice in RLE PACKED
decode to reduce allocation overhead
3d2b220ee59b is described below
commit 3d2b220ee59ba2bf211f949bc4e7b2938854a088
Author: Ismaël Mejía <[email protected]>
AuthorDate: Tue Jun 23 17:29:05 2026 +0200
[SPARK-56895][SQL] Batch ByteBuffer slice in RLE PACKED decode to reduce
allocation overhead
### What changes were proposed in this pull request?
In `VectorizedRleValuesReader.readNextGroup()`, the PACKED branch
previously called `in.slice(bitWidth)` once per group of 8 values, allocating a
new `ByteBuffer` view per call. This PR batches all packed bytes into a single
`in.slice(numGroups * bitWidth)` call, then indexes into the resulting buffer
with an advancing position offset.
The change is 4 lines of net diff in `readNextGroup()`:
```java
// Before: one slice() per group — N ByteBuffer allocations
while (valueIndex < this.currentCount) {
ByteBuffer buffer = in.slice(bitWidth);
this.packer.unpack8Values(buffer, buffer.position(),
this.currentBuffer, valueIndex);
valueIndex += 8;
}
// After: single slice() for all groups — 1 ByteBuffer allocation
int totalBytes = numGroups * bitWidth;
ByteBuffer packed = in.slice(totalBytes);
int pos = packed.position();
while (valueIndex < this.currentCount) {
this.packer.unpack8Values(packed, pos, this.currentBuffer, valueIndex);
pos += bitWidth;
valueIndex += 8;
}
```
This works because `BytePacker.unpack8Values(ByteBuffer, int inPos, int[],
int)` uses absolute positioning (`inPos` is the absolute offset into the
buffer, not relative to `position()`).
### Why are the changes needed?
`ByteBufferInputStream.slice()` calls `buffer.duplicate()` which allocates
a new `ByteBuffer` object each time. For a 1M-value PACKED page at bitWidth=4,
that is 128K `ByteBuffer` allocations per page. These short-lived objects
create GC pressure and prevent `BytePacker.unpack8Values()` from being inlined
(the `ByteBuffer` argument escapes through the abstract method call, defeating
scalar replacement).
With a single slice, the entire PACKED block shares one `ByteBuffer` view.
The `unpack8Values` calls index into it via an advancing absolute position,
which C2 can optimize far more aggressively.
Benchmark results on the same machine (AMD EPYC 9V45, JDK 17), before vs
after:
**readIntegers (dictionary-id decode, PACKED):**
| bitWidth | Before (M/s) | After (M/s) | Speedup |
|----------|-------------|-------------|---------|
| 4 | 886 | 1862 | 2.1x |
| 8 | 885 | 2088 | 2.4x |
| 12 | 609 | 953 | 1.6x |
| 20 | 518 | 709 | 1.4x |
**Single-value reads (readInteger, PACKED):**
| bitWidth | Before (M/s) | After (M/s) | Speedup |
|----------|-------------|-------------|---------|
| 4 | 490 | 690 | 1.4x |
| 8 | 495 | 722 | 1.5x |
| 12 | 392 | 510 | 1.3x |
| 20 | 354 | 431 | 1.2x |
**skipIntegers (PACKED):**
| bitWidth | Before (M/s) | After (M/s) | Speedup |
|----------|-------------|-------------|---------|
| 4 | 916 | 1942 | 2.1x |
| 8 | 915 | 2238 | 2.4x |
| 12 | 622 | 988 | 1.6x |
| 20 | 528 | 722 | 1.4x |
RLE paths are unaffected (no PACKED decode involved). The benefit is
largest at smaller bitWidths where more groups (and therefore more slice calls)
are needed per page.
GHA results on AMD EPYC 7763 (PACKED reused, readIntegers):
| bitWidth | JDK 17 Baseline | JDK 17 PR | Speedup | JDK 21 Baseline | JDK
21 PR | Speedup | JDK 25 Baseline | JDK 25 PR | Speedup |
|---|---|---|---|---|---|---|---|---|---|
| 4 | 485.2 | 980.3 | **2.0x** | 541.2 | 897.1 | **1.7x** | 522.1 | 891.0 |
**1.7x** |
| 8 | 476.5 | 949.7 | **2.0x** | 618.1 | 1006.4 | **1.6x** | 579.4 | 1049.4
| **1.8x** |
| 12 | 357.3 | 563.8 | **1.6x** | 480.5 | 681.5 | **1.4x** | 436.4 | 689.4
| **1.6x** |
| 20 | 302.1 | 441.8 | **1.5x** | 400.6 | 522.8 | **1.3x** | 382.8 | 520.2
| **1.4x** |
Full committed results: [JDK
17](https://github.com/iemejia/spark/actions/runs/27443135287), [JDK
21](https://github.com/iemejia/spark/actions/runs/27435379671), [JDK
25](https://github.com/iemejia/spark/actions/runs/27437123091)
### Does this PR introduce _any_ user-facing change?
No. This is an internal performance optimization in the Parquet vectorized
reader. No API, configuration, or behavioral changes.
### How was this patch tested?
- Existing `VectorizedRleValuesReaderSuite` (14 tests covering PACKED mode,
RLE mode, mixed, row-index filtering, multi-page, cross-batch continuity) — all
pass.
- Existing `ParquetQuerySuite`, `ParquetIOSuite`, `ParquetEncodingSuite`,
`ParquetSchemaSuite` — all 244 tests pass on JDK 17.
- `VectorizedRleValuesReaderBenchmark` — before/after comparison on the
same machine shown above.
### Was this patch authored or co-authored using generative AI tooling?
Generated-by: OpenCode (Claude claude-opus-4.6)
Closes #55922 from iemejia/SPARK-56895-rle-packed-batch-slice.
Authored-by: Ismaël Mejía <[email protected]>
Signed-off-by: Max Gekk <[email protected]>
(cherry picked from commit acc16fb244a86d30a9d0e96c0943fae6b8fcfc42)
Signed-off-by: Max Gekk <[email protected]>
---
...rizedRleValuesReaderBenchmark-jdk21-results.txt | 180 ++++++++++-----------
...rizedRleValuesReaderBenchmark-jdk25-results.txt | 180 ++++++++++-----------
.../VectorizedRleValuesReaderBenchmark-results.txt | 156 +++++++++---------
.../parquet/VectorizedRleValuesReader.java | 10 +-
.../parquet/VectorizedRleValuesReaderSuite.scala | 57 +++++++
5 files changed, 322 insertions(+), 261 deletions(-)
diff --git
a/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-jdk21-results.txt
b/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-jdk21-results.txt
index b5f6d62d2b6b..1d8f621294d1 100644
--- a/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-jdk21-results.txt
+++ b/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-jdk21-results.txt
@@ -2,153 +2,153 @@
Boolean decode
================================================================================================
-OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1013-azure
-AMD EPYC 9V74 80-Core Processor
+OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
RLE readBooleans decode: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-cold reader, trueRatio=0.0 0 0
0 4886.6 0.2 1.0X
-reused reader, trueRatio=0.0 0 0
0 4857.7 0.2 1.0X
-cold reader, trueRatio=0.1 1 1
0 1186.3 0.8 0.2X
-reused reader, trueRatio=0.1 1 1
0 789.7 1.3 0.2X
-cold reader, trueRatio=0.5 1 1
0 1335.3 0.7 0.3X
-reused reader, trueRatio=0.5 1 1
0 855.2 1.2 0.2X
-cold reader, trueRatio=0.9 1 1
0 1186.3 0.8 0.2X
-reused reader, trueRatio=0.9 1 1
0 787.6 1.3 0.2X
-cold reader, trueRatio=1.0 0 0
0 4064.6 0.2 0.8X
-reused reader, trueRatio=1.0 0 0
0 4855.5 0.2 1.0X
+cold reader, trueRatio=0.0 0 0
0 12677.1 0.1 1.0X
+reused reader, trueRatio=0.0 0 0
0 12680.2 0.1 1.0X
+cold reader, trueRatio=0.1 2 2
0 694.6 1.4 0.1X
+reused reader, trueRatio=0.1 2 2
0 615.1 1.6 0.0X
+cold reader, trueRatio=0.5 1 1
0 734.1 1.4 0.1X
+reused reader, trueRatio=0.5 2 2
0 653.3 1.5 0.1X
+cold reader, trueRatio=0.9 2 2
0 693.8 1.4 0.1X
+reused reader, trueRatio=0.9 2 2
0 613.0 1.6 0.0X
+cold reader, trueRatio=1.0 0 0
0 12675.4 0.1 1.0X
+reused reader, trueRatio=1.0 0 0
0 12678.7 0.1 1.0X
================================================================================================
Integer decode
================================================================================================
-OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1013-azure
-AMD EPYC 9V74 80-Core Processor
+OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
RLE readIntegers dictionary-id decode: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-PACKED cold, bitWidth=4 2 2
0 542.4 1.8 1.0X
-PACKED reused, bitWidth=4 2 2
0 541.2 1.8 1.0X
-RLE, bitWidth=4 0 0
0 23543.9 0.0 43.4X
-PACKED cold, bitWidth=8 2 2
0 621.3 1.6 1.1X
-PACKED reused, bitWidth=8 2 2
0 618.1 1.6 1.1X
-RLE, bitWidth=8 0 0
0 23511.7 0.0 43.3X
-PACKED cold, bitWidth=12 2 2
0 482.5 2.1 0.9X
-PACKED reused, bitWidth=12 2 2
0 480.5 2.1 0.9X
-RLE, bitWidth=12 0 0
0 23507.0 0.0 43.3X
-PACKED cold, bitWidth=20 3 3
0 401.9 2.5 0.7X
-PACKED reused, bitWidth=20 3 3
0 400.6 2.5 0.7X
-RLE, bitWidth=20 0 0
0 23570.4 0.0 43.5X
+PACKED cold, bitWidth=4 1 1
0 900.8 1.1 1.0X
+PACKED reused, bitWidth=4 1 1
0 897.1 1.1 1.0X
+RLE, bitWidth=4 0 0
0 15764.7 0.1 17.5X
+PACKED cold, bitWidth=8 1 1
0 1033.6 1.0 1.1X
+PACKED reused, bitWidth=8 1 1
0 1006.4 1.0 1.1X
+RLE, bitWidth=8 0 0
0 20259.6 0.0 22.5X
+PACKED cold, bitWidth=12 2 2
0 684.1 1.5 0.8X
+PACKED reused, bitWidth=12 2 2
0 681.5 1.5 0.8X
+RLE, bitWidth=12 0 0
0 20220.5 0.0 22.4X
+PACKED cold, bitWidth=20 2 2
0 523.5 1.9 0.6X
+PACKED reused, bitWidth=20 2 2
0 522.8 1.9 0.6X
+RLE, bitWidth=20 0 0
0 15767.1 0.1 17.5X
================================================================================================
Nullable batch decode with def-level materialization
================================================================================================
-OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1013-azure
-AMD EPYC 9V74 80-Core Processor
+OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Nullable batch with def-levels: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, n/a 0 0
0 8154.1 0.1 1.0X
-nullRatio=0.1, random 7 8
0 140.5 7.1 0.0X
-nullRatio=0.1, clustered 5 5
0 204.1 4.9 0.0X
-nullRatio=0.3, random 11 11
0 96.0 10.4 0.0X
-nullRatio=0.3, clustered 5 5
0 207.8 4.8 0.0X
-nullRatio=0.5, random 12 12
0 87.4 11.4 0.0X
-nullRatio=0.5, clustered 5 5
0 213.6 4.7 0.0X
-nullRatio=0.9, random 6 7
0 162.1 6.2 0.0X
-nullRatio=0.9, clustered 4 5
0 235.1 4.3 0.0X
-nullRatio=1.0, random 0 0
0 22458.3 0.0 2.8X
+nullRatio=0.0, n/a 0 0
0 6620.0 0.2 1.0X
+nullRatio=0.1, random 8 8
0 132.9 7.5 0.0X
+nullRatio=0.1, clustered 5 5
0 193.5 5.2 0.0X
+nullRatio=0.3, random 12 12
0 91.0 11.0 0.0X
+nullRatio=0.3, clustered 5 5
0 197.2 5.1 0.0X
+nullRatio=0.5, random 13 13
0 81.2 12.3 0.0X
+nullRatio=0.5, clustered 5 5
0 203.5 4.9 0.0X
+nullRatio=0.9, random 7 7
0 155.1 6.4 0.0X
+nullRatio=0.9, clustered 5 5
0 227.8 4.4 0.0X
+nullRatio=1.0, random 0 0
0 8305.8 0.1 1.3X
================================================================================================
Nullable batch decode without def-level materialization
================================================================================================
-OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1013-azure
-AMD EPYC 9V74 80-Core Processor
+OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Nullable batch without def-levels: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, n/a 0 0
0 12686.2 0.1 1.0X
-nullRatio=0.1, random 6 6
0 172.3 5.8 0.0X
-nullRatio=0.1, clustered 4 4
0 251.2 4.0 0.0X
-nullRatio=0.3, random 9 9
0 115.5 8.7 0.0X
-nullRatio=0.3, clustered 4 4
0 253.9 3.9 0.0X
-nullRatio=0.5, random 10 10
0 105.1 9.5 0.0X
-nullRatio=0.5, clustered 4 4
0 259.7 3.9 0.0X
-nullRatio=0.9, random 5 5
0 198.9 5.0 0.0X
-nullRatio=0.9, clustered 4 4
0 282.6 3.5 0.0X
-nullRatio=1.0, random 0 0
0 96058.6 0.0 7.6X
+nullRatio=0.0, n/a 0 0
0 11471.1 0.1 1.0X
+nullRatio=0.1, random 6 6
0 163.2 6.1 0.0X
+nullRatio=0.1, clustered 4 4
0 244.7 4.1 0.0X
+nullRatio=0.3, random 10 10
0 109.6 9.1 0.0X
+nullRatio=0.3, clustered 4 4
0 246.4 4.1 0.0X
+nullRatio=0.5, random 11 11
0 97.2 10.3 0.0X
+nullRatio=0.5, clustered 4 4
0 253.4 3.9 0.0X
+nullRatio=0.9, random 6 6
0 190.1 5.3 0.0X
+nullRatio=0.9, clustered 4 4
0 283.7 3.5 0.0X
+nullRatio=1.0, random 0 0
0 12006.8 0.1 1.0X
================================================================================================
Nullable batch decode with row-index filtering (with def-levels)
================================================================================================
-OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1013-azure
-AMD EPYC 9V74 80-Core Processor
+OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Nullable batch with def-levels, row-index filtered: Best Time(ms) Avg
Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
----------------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, contiguous 50% 1
1 0 1247.7 0.8 1.0X
-nullRatio=0.3, contiguous 50% 8
8 0 135.1 7.4 0.1X
-nullRatio=0.9, contiguous 50% 5
6 0 191.0 5.2 0.2X
-nullRatio=0.0, alt 1000-row windows 2
2 0 433.2 2.3 0.3X
-nullRatio=0.3, alt 1000-row windows 9
9 0 113.6 8.8 0.1X
-nullRatio=0.9, alt 1000-row windows 7
7 0 150.4 6.6 0.1X
+nullRatio=0.0, contiguous 50% 1
1 0 765.3 1.3 1.0X
+nullRatio=0.3, contiguous 50% 8
9 0 125.3 8.0 0.2X
+nullRatio=0.9, contiguous 50% 6
6 0 181.6 5.5 0.2X
+nullRatio=0.0, alt 1000-row windows 3
3 0 378.3 2.6 0.5X
+nullRatio=0.3, alt 1000-row windows 10
10 0 106.8 9.4 0.1X
+nullRatio=0.9, alt 1000-row windows 7
7 0 145.5 6.9 0.2X
================================================================================================
Nullable batch decode with row-index filtering (without def-levels)
================================================================================================
-OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1013-azure
-AMD EPYC 9V74 80-Core Processor
+OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Nullable batch without def-levels, row-index filtered: Best Time(ms) Avg
Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
-------------------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, contiguous 50% 1
1 0 1004.6 1.0 1.0X
-nullRatio=0.3, contiguous 50% 7
7 1 154.7 6.5 0.2X
-nullRatio=0.9, contiguous 50% 5
5 0 214.0 4.7 0.2X
-nullRatio=0.0, alt 1000-row windows 2
2 0 464.0 2.2 0.5X
-nullRatio=0.3, alt 1000-row windows 8
8 0 130.3 7.7 0.1X
-nullRatio=0.9, alt 1000-row windows 6
6 0 169.6 5.9 0.2X
+nullRatio=0.0, contiguous 50% 1
4 1 749.1 1.3 1.0X
+nullRatio=0.3, contiguous 50% 7
7 0 146.3 6.8 0.2X
+nullRatio=0.9, contiguous 50% 5
5 0 203.9 4.9 0.3X
+nullRatio=0.0, alt 1000-row windows 3
3 0 382.9 2.6 0.5X
+nullRatio=0.3, alt 1000-row windows 9
9 0 121.7 8.2 0.2X
+nullRatio=0.9, alt 1000-row windows 7
7 0 159.7 6.3 0.2X
================================================================================================
Single-value reads
================================================================================================
-OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1013-azure
-AMD EPYC 9V74 80-Core Processor
+OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Single-value reads: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-readBoolean 3 3
0 361.8 2.8 1.0X
-readInteger, bitWidth=4 3 3
0 314.3 3.2 0.9X
-readValueDictionaryId, bitWidth=4 3 3
0 315.3 3.2 0.9X
-readInteger, bitWidth=8 3 3
0 339.9 2.9 0.9X
-readValueDictionaryId, bitWidth=8 3 3
0 339.8 2.9 0.9X
-readInteger, bitWidth=12 4 4
0 293.8 3.4 0.8X
-readValueDictionaryId, bitWidth=12 4 4
0 293.9 3.4 0.8X
-readInteger, bitWidth=20 4 4
0 262.1 3.8 0.7X
-readValueDictionaryId, bitWidth=20 4 4
0 261.9 3.8 0.7X
+readBoolean 2 2
0 441.2 2.3 1.0X
+readInteger, bitWidth=4 3 3
0 368.7 2.7 0.8X
+readValueDictionaryId, bitWidth=4 3 3
0 368.9 2.7 0.8X
+readInteger, bitWidth=8 3 3
0 393.7 2.5 0.9X
+readValueDictionaryId, bitWidth=8 3 3
0 394.1 2.5 0.9X
+readInteger, bitWidth=12 3 3
0 330.7 3.0 0.7X
+readValueDictionaryId, bitWidth=12 3 3
0 329.7 3.0 0.7X
+readInteger, bitWidth=20 4 4
0 288.4 3.5 0.7X
+readValueDictionaryId, bitWidth=20 4 4
0 287.8 3.5 0.7X
================================================================================================
Skip
================================================================================================
-OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1013-azure
-AMD EPYC 9V74 80-Core Processor
+OpenJDK 64-Bit Server VM 21.0.11+10-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Skip: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-skipBooleans, trueRatio=0.0 0 0
0 34952533.3 0.0 1.0X
-skipBooleans, trueRatio=0.5 2 2
0 662.2 1.5 0.0X
-skipBooleans, trueRatio=1.0 0 0
0 34952533.3 0.0 1.0X
-skipIntegers PACKED, bitWidth=4 2 2
0 553.8 1.8 0.0X
-skipIntegers RLE, bitWidth=4 0 0
0 34952533.3 0.0 1.0X
-skipIntegers PACKED, bitWidth=8 2 2
0 637.9 1.6 0.0X
-skipIntegers RLE, bitWidth=8 0 0
0 34952533.3 0.0 1.0X
-skipIntegers PACKED, bitWidth=12 2 2
0 493.4 2.0 0.0X
-skipIntegers RLE, bitWidth=12 0 0
0 26214400.0 0.0 0.8X
-skipIntegers PACKED, bitWidth=20 3 3
0 415.6 2.4 0.0X
-skipIntegers RLE, bitWidth=20 0 0
0 26214400.0 0.0 0.8X
+skipBooleans, trueRatio=0.0 0 0
0 21399510.2 0.0 1.0X
+skipBooleans, trueRatio=0.5 1 1
0 1184.1 0.8 0.0X
+skipBooleans, trueRatio=1.0 0 0
0 21399510.2 0.0 1.0X
+skipIntegers PACKED, bitWidth=4 1 1
0 924.2 1.1 0.0X
+skipIntegers RLE, bitWidth=4 0 0
0 21399510.2 0.0 1.0X
+skipIntegers PACKED, bitWidth=8 1 1
0 1117.6 0.9 0.0X
+skipIntegers RLE, bitWidth=8 0 0
0 21399510.2 0.0 1.0X
+skipIntegers PACKED, bitWidth=12 1 1
0 720.0 1.4 0.0X
+skipIntegers RLE, bitWidth=12 0 0
0 21399510.2 0.0 1.0X
+skipIntegers PACKED, bitWidth=20 2 2
0 546.7 1.8 0.0X
+skipIntegers RLE, bitWidth=20 0 0
0 21399510.2 0.0 1.0X
diff --git
a/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-jdk25-results.txt
b/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-jdk25-results.txt
index 860fafe62b1e..4e9d6791cde2 100644
--- a/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-jdk25-results.txt
+++ b/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-jdk25-results.txt
@@ -2,153 +2,153 @@
Boolean decode
================================================================================================
-OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1013-azure
-Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz
+OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
RLE readBooleans decode: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-cold reader, trueRatio=0.0 0 0
0 64903.2 0.0 1.0X
-reused reader, trueRatio=0.0 0 0
0 63220.5 0.0 1.0X
-cold reader, trueRatio=0.1 1 1
0 830.6 1.2 0.0X
-reused reader, trueRatio=0.1 1 1
0 789.9 1.3 0.0X
-cold reader, trueRatio=0.5 1 1
0 927.6 1.1 0.0X
-reused reader, trueRatio=0.5 1 1
0 905.4 1.1 0.0X
-cold reader, trueRatio=0.9 1 1
0 831.8 1.2 0.0X
-reused reader, trueRatio=0.9 1 1
0 833.2 1.2 0.0X
-cold reader, trueRatio=1.0 0 0
0 64176.3 0.0 1.0X
-reused reader, trueRatio=1.0 0 0
0 62706.4 0.0 1.0X
+cold reader, trueRatio=0.0 0 0
0 62785.2 0.0 1.0X
+reused reader, trueRatio=0.0 0 0
0 82280.0 0.0 1.3X
+cold reader, trueRatio=0.1 2 2
0 662.8 1.5 0.0X
+reused reader, trueRatio=0.1 2 2
0 655.5 1.5 0.0X
+cold reader, trueRatio=0.5 1 2
0 710.3 1.4 0.0X
+reused reader, trueRatio=0.5 1 2
0 700.6 1.4 0.0X
+cold reader, trueRatio=0.9 2 2
0 662.2 1.5 0.0X
+reused reader, trueRatio=0.9 2 2
0 655.1 1.5 0.0X
+cold reader, trueRatio=1.0 0 0
0 82280.0 0.0 1.3X
+reused reader, trueRatio=1.0 0 0
0 62560.5 0.0 1.0X
================================================================================================
Integer decode
================================================================================================
-OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1013-azure
-Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz
+OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
RLE readIntegers dictionary-id decode: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-PACKED cold, bitWidth=4 2 2
0 534.3 1.9 1.0X
-PACKED reused, bitWidth=4 2 2
0 522.1 1.9 1.0X
-RLE, bitWidth=4 0 0
0 7632.3 0.1 14.3X
-PACKED cold, bitWidth=8 2 2
0 584.0 1.7 1.1X
-PACKED reused, bitWidth=8 2 2
0 579.4 1.7 1.1X
-RLE, bitWidth=8 0 0
0 7615.7 0.1 14.3X
-PACKED cold, bitWidth=12 2 2
0 443.8 2.3 0.8X
-PACKED reused, bitWidth=12 2 3
0 436.4 2.3 0.8X
-RLE, bitWidth=12 0 0
0 7589.5 0.1 14.2X
-PACKED cold, bitWidth=20 3 3
0 378.5 2.6 0.7X
-PACKED reused, bitWidth=20 3 3
0 382.8 2.6 0.7X
-RLE, bitWidth=20 0 0
0 7590.5 0.1 14.2X
+PACKED cold, bitWidth=4 1 1
0 896.2 1.1 1.0X
+PACKED reused, bitWidth=4 1 1
0 891.0 1.1 1.0X
+RLE, bitWidth=4 0 0
0 15783.5 0.1 17.6X
+PACKED cold, bitWidth=8 1 1
0 1065.8 0.9 1.2X
+PACKED reused, bitWidth=8 1 1
0 1049.4 1.0 1.2X
+RLE, bitWidth=8 0 0
0 15780.9 0.1 17.6X
+PACKED cold, bitWidth=12 2 2
0 693.7 1.4 0.8X
+PACKED reused, bitWidth=12 2 2
0 689.4 1.5 0.8X
+RLE, bitWidth=12 0 0
0 15778.5 0.1 17.6X
+PACKED cold, bitWidth=20 2 2
0 537.3 1.9 0.6X
+PACKED reused, bitWidth=20 2 2
0 520.2 1.9 0.6X
+RLE, bitWidth=20 0 0
0 15783.5 0.1 17.6X
================================================================================================
Nullable batch decode with def-level materialization
================================================================================================
-OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1013-azure
-Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz
+OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Nullable batch with def-levels: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, n/a 0 0
0 5285.1 0.2 1.0X
-nullRatio=0.1, random 9 9
0 113.1 8.8 0.0X
-nullRatio=0.1, clustered 6 6
0 173.0 5.8 0.0X
-nullRatio=0.3, random 14 14
0 75.5 13.2 0.0X
-nullRatio=0.3, clustered 6 6
0 173.1 5.8 0.0X
-nullRatio=0.5, random 15 16
0 67.7 14.8 0.0X
-nullRatio=0.5, clustered 6 6
0 176.2 5.7 0.0X
-nullRatio=0.9, random 8 8
0 128.6 7.8 0.0X
-nullRatio=0.9, clustered 5 6
0 196.2 5.1 0.0X
-nullRatio=1.0, random 0 0
0 35936.0 0.0 6.8X
+nullRatio=0.0, n/a 0 0
0 6836.1 0.1 1.0X
+nullRatio=0.1, random 8 8
0 137.1 7.3 0.0X
+nullRatio=0.1, clustered 5 5
0 203.3 4.9 0.0X
+nullRatio=0.3, random 11 11
0 92.8 10.8 0.0X
+nullRatio=0.3, clustered 5 5
0 205.3 4.9 0.0X
+nullRatio=0.5, random 13 13
0 82.3 12.1 0.0X
+nullRatio=0.5, clustered 5 5
0 209.5 4.8 0.0X
+nullRatio=0.9, random 7 7
1 154.8 6.5 0.0X
+nullRatio=0.9, clustered 5 5
0 230.2 4.3 0.0X
+nullRatio=1.0, random 0 0
0 19063.6 0.1 2.8X
================================================================================================
Nullable batch decode without def-level materialization
================================================================================================
-OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1013-azure
-Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz
+OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Nullable batch without def-levels: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, n/a 0 0
0 8150.4 0.1 1.0X
-nullRatio=0.1, random 7 7
0 147.0 6.8 0.0X
-nullRatio=0.1, clustered 5 5
0 218.2 4.6 0.0X
-nullRatio=0.3, random 11 11
0 98.0 10.2 0.0X
-nullRatio=0.3, clustered 5 5
1 218.0 4.6 0.0X
-nullRatio=0.5, random 12 12
0 90.0 11.1 0.0X
-nullRatio=0.5, clustered 5 5
0 221.1 4.5 0.0X
-nullRatio=0.9, random 6 6
0 167.0 6.0 0.0X
-nullRatio=0.9, clustered 4 5
0 237.7 4.2 0.0X
-nullRatio=1.0, random 0 0
0 115647.5 0.0 14.2X
+nullRatio=0.0, n/a 0 0
0 12591.4 0.1 1.0X
+nullRatio=0.1, random 6 6
0 162.5 6.2 0.0X
+nullRatio=0.1, clustered 4 4
0 253.5 3.9 0.0X
+nullRatio=0.3, random 9 10
1 111.1 9.0 0.0X
+nullRatio=0.3, clustered 4 4
0 253.6 3.9 0.0X
+nullRatio=0.5, random 11 11
0 98.8 10.1 0.0X
+nullRatio=0.5, clustered 4 4
0 257.7 3.9 0.0X
+nullRatio=0.9, random 6 6
1 186.5 5.4 0.0X
+nullRatio=0.9, clustered 4 4
0 280.6 3.6 0.0X
+nullRatio=1.0, random 0 0
0 66453.9 0.0 5.3X
================================================================================================
Nullable batch decode with row-index filtering (with def-levels)
================================================================================================
-OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1013-azure
-Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz
+OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Nullable batch with def-levels, row-index filtered: Best Time(ms) Avg
Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
----------------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, contiguous 50% 1
1 0 817.3 1.2 1.0X
-nullRatio=0.3, contiguous 50% 9
10 0 111.1 9.0 0.1X
-nullRatio=0.9, contiguous 50% 7
7 0 159.5 6.3 0.2X
-nullRatio=0.0, alt 1000-row windows 3
3 0 327.9 3.1 0.4X
-nullRatio=0.3, alt 1000-row windows 12
12 1 90.9 11.0 0.1X
-nullRatio=0.9, alt 1000-row windows 9
9 0 119.7 8.4 0.1X
+nullRatio=0.0, contiguous 50% 1
2 0 765.3 1.3 1.0X
+nullRatio=0.3, contiguous 50% 8
8 0 134.3 7.4 0.2X
+nullRatio=0.9, contiguous 50% 6
6 0 188.3 5.3 0.2X
+nullRatio=0.0, alt 1000-row windows 2
2 0 435.3 2.3 0.6X
+nullRatio=0.3, alt 1000-row windows 9
9 0 118.4 8.4 0.2X
+nullRatio=0.9, alt 1000-row windows 7
7 0 159.2 6.3 0.2X
================================================================================================
Nullable batch decode with row-index filtering (without def-levels)
================================================================================================
-OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1013-azure
-Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz
+OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Nullable batch without def-levels, row-index filtered: Best Time(ms) Avg
Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
-------------------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, contiguous 50% 1
1 0 842.7 1.2 1.0X
-nullRatio=0.3, contiguous 50% 8
8 0 128.2 7.8 0.2X
-nullRatio=0.9, contiguous 50% 6
6 0 181.8 5.5 0.2X
-nullRatio=0.0, alt 1000-row windows 3
3 0 331.8 3.0 0.4X
-nullRatio=0.3, alt 1000-row windows 10
11 0 102.5 9.8 0.1X
-nullRatio=0.9, alt 1000-row windows 8
8 0 134.2 7.4 0.2X
+nullRatio=0.0, contiguous 50% 1
2 0 736.9 1.4 1.0X
+nullRatio=0.3, contiguous 50% 7
7 0 143.6 7.0 0.2X
+nullRatio=0.9, contiguous 50% 5
5 0 197.3 5.1 0.3X
+nullRatio=0.0, alt 1000-row windows 2
2 0 442.1 2.3 0.6X
+nullRatio=0.3, alt 1000-row windows 8
8 0 126.8 7.9 0.2X
+nullRatio=0.9, alt 1000-row windows 6
6 0 167.1 6.0 0.2X
================================================================================================
Single-value reads
================================================================================================
-OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1013-azure
-Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz
+OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Single-value reads: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-readBoolean 4 4
0 263.6 3.8 1.0X
-readInteger, bitWidth=4 3 4
0 301.6 3.3 1.1X
-readValueDictionaryId, bitWidth=4 3 4
0 300.4 3.3 1.1X
-readInteger, bitWidth=8 3 4
0 314.9 3.2 1.2X
-readValueDictionaryId, bitWidth=8 3 3
0 315.7 3.2 1.2X
-readInteger, bitWidth=12 4 4
0 276.9 3.6 1.1X
-readValueDictionaryId, bitWidth=12 4 4
0 275.9 3.6 1.0X
-readInteger, bitWidth=20 4 4
0 249.3 4.0 0.9X
-readValueDictionaryId, bitWidth=20 4 4
0 247.9 4.0 0.9X
+readBoolean 2 2
0 427.5 2.3 1.0X
+readInteger, bitWidth=4 3 3
0 404.5 2.5 0.9X
+readValueDictionaryId, bitWidth=4 3 3
0 404.5 2.5 0.9X
+readInteger, bitWidth=8 2 2
0 440.5 2.3 1.0X
+readValueDictionaryId, bitWidth=8 2 2
0 440.3 2.3 1.0X
+readInteger, bitWidth=12 3 3
0 359.8 2.8 0.8X
+readValueDictionaryId, bitWidth=12 3 3
0 359.5 2.8 0.8X
+readInteger, bitWidth=20 3 3
0 314.1 3.2 0.7X
+readValueDictionaryId, bitWidth=20 3 3
0 314.1 3.2 0.7X
================================================================================================
Skip
================================================================================================
-OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1013-azure
-Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz
+OpenJDK 64-Bit Server VM 25.0.3+9-LTS on Linux 6.17.0-1018-azure
+AMD EPYC 7763 64-Core Processor
Skip: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-skipBooleans, trueRatio=0.0 0 0
0 29959314.3 0.0 1.0X
-skipBooleans, trueRatio=0.5 2 2
0 611.9 1.6 0.0X
-skipBooleans, trueRatio=1.0 0 0
0 29959314.3 0.0 1.0X
-skipIntegers PACKED, bitWidth=4 2 2
0 542.6 1.8 0.0X
-skipIntegers RLE, bitWidth=4 0 0
0 29959314.3 0.0 1.0X
-skipIntegers PACKED, bitWidth=8 2 2
0 590.8 1.7 0.0X
-skipIntegers RLE, bitWidth=8 0 0
0 29959314.3 0.0 1.0X
-skipIntegers PACKED, bitWidth=12 2 2
0 467.9 2.1 0.0X
-skipIntegers RLE, bitWidth=12 0 0
0 26214400.0 0.0 0.9X
-skipIntegers PACKED, bitWidth=20 3 3
0 404.2 2.5 0.0X
-skipIntegers RLE, bitWidth=20 0 0
0 23831272.7 0.0 0.8X
+skipBooleans, trueRatio=0.0 0 0
0 26214400.0 0.0 1.0X
+skipBooleans, trueRatio=0.5 1 1
0 1192.5 0.8 0.0X
+skipBooleans, trueRatio=1.0 0 0
0 26214400.0 0.0 1.0X
+skipIntegers PACKED, bitWidth=4 1 1
0 944.1 1.1 0.0X
+skipIntegers RLE, bitWidth=4 0 0
0 26214400.0 0.0 1.0X
+skipIntegers PACKED, bitWidth=8 1 1
0 1175.4 0.9 0.0X
+skipIntegers RLE, bitWidth=8 0 0
0 26214400.0 0.0 1.0X
+skipIntegers PACKED, bitWidth=12 1 1
0 733.2 1.4 0.0X
+skipIntegers RLE, bitWidth=12 0 0
0 21399510.2 0.0 0.8X
+skipIntegers PACKED, bitWidth=20 2 2
0 560.2 1.8 0.0X
+skipIntegers RLE, bitWidth=20 0 0
0 21399510.2 0.0 0.8X
diff --git a/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-results.txt
b/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-results.txt
index 2ec7742a9d13..e694026f0dc8 100644
--- a/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-results.txt
+++ b/sql/core/benchmarks/VectorizedRleValuesReaderBenchmark-results.txt
@@ -2,153 +2,153 @@
Boolean decode
================================================================================================
-OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1013-azure
+OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 7763 64-Core Processor
RLE readBooleans decode: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-cold reader, trueRatio=0.0 0 0
0 25842.3 0.0 1.0X
-reused reader, trueRatio=0.0 0 0
0 25810.5 0.0 1.0X
-cold reader, trueRatio=0.1 2 2
0 485.6 2.1 0.0X
-reused reader, trueRatio=0.1 2 2
0 485.6 2.1 0.0X
-cold reader, trueRatio=0.5 2 2
0 500.3 2.0 0.0X
-reused reader, trueRatio=0.5 2 2
0 483.7 2.1 0.0X
-cold reader, trueRatio=0.9 2 2
0 484.3 2.1 0.0X
-reused reader, trueRatio=0.9 2 2
0 484.7 2.1 0.0X
-cold reader, trueRatio=1.0 0 0
0 25804.1 0.0 1.0X
-reused reader, trueRatio=1.0 0 0
0 25791.4 0.0 1.0X
+cold reader, trueRatio=0.0 0 0
0 25855.0 0.0 1.0X
+reused reader, trueRatio=0.0 0 0
0 25855.0 0.0 1.0X
+cold reader, trueRatio=0.1 1 1
0 1047.4 1.0 0.0X
+reused reader, trueRatio=0.1 1 1
0 1045.0 1.0 0.0X
+cold reader, trueRatio=0.5 1 1
0 1268.1 0.8 0.0X
+reused reader, trueRatio=0.5 1 1
0 1271.9 0.8 0.0X
+cold reader, trueRatio=0.9 1 1
0 1045.7 1.0 0.0X
+reused reader, trueRatio=0.9 1 1
0 1044.1 1.0 0.0X
+cold reader, trueRatio=1.0 0 0
0 25848.0 0.0 1.0X
+reused reader, trueRatio=1.0 0 0
0 25861.4 0.0 1.0X
================================================================================================
Integer decode
================================================================================================
-OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1013-azure
+OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 7763 64-Core Processor
RLE readIntegers dictionary-id decode: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-PACKED cold, bitWidth=4 2 2
0 485.7 2.1 1.0X
-PACKED reused, bitWidth=4 2 2
0 485.2 2.1 1.0X
-RLE, bitWidth=4 0 0
0 20688.1 0.0 42.6X
-PACKED cold, bitWidth=8 2 2
0 478.8 2.1 1.0X
-PACKED reused, bitWidth=8 2 2
0 476.5 2.1 1.0X
-RLE, bitWidth=8 0 0
0 15710.2 0.1 32.3X
-PACKED cold, bitWidth=12 3 3
0 357.4 2.8 0.7X
-PACKED reused, bitWidth=12 3 3
0 357.3 2.8 0.7X
-RLE, bitWidth=12 0 0
0 20684.0 0.0 42.6X
-PACKED cold, bitWidth=20 3 4
0 303.0 3.3 0.6X
-PACKED reused, bitWidth=20 3 4
0 302.1 3.3 0.6X
-RLE, bitWidth=20 0 0
0 20675.9 0.0 42.6X
+PACKED cold, bitWidth=4 1 1
0 984.5 1.0 1.0X
+PACKED reused, bitWidth=4 1 1
0 980.3 1.0 1.0X
+RLE, bitWidth=4 0 0
0 20824.1 0.0 21.2X
+PACKED cold, bitWidth=8 1 1
0 957.3 1.0 1.0X
+PACKED reused, bitWidth=8 1 1
0 949.7 1.1 1.0X
+RLE, bitWidth=8 0 0
0 20757.7 0.0 21.1X
+PACKED cold, bitWidth=12 2 2
0 566.9 1.8 0.6X
+PACKED reused, bitWidth=12 2 2
0 563.8 1.8 0.6X
+RLE, bitWidth=12 0 0
0 20189.0 0.0 20.5X
+PACKED cold, bitWidth=20 2 2
0 448.7 2.2 0.5X
+PACKED reused, bitWidth=20 2 2
0 441.8 2.3 0.4X
+RLE, bitWidth=20 0 0
0 20201.1 0.0 20.5X
================================================================================================
Nullable batch decode with def-level materialization
================================================================================================
-OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1013-azure
+OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 7763 64-Core Processor
Nullable batch with def-levels: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, n/a 0 0
0 6435.2 0.2 1.0X
-nullRatio=0.1, random 10 10
0 106.0 9.4 0.0X
-nullRatio=0.1, clustered 7 7
0 144.9 6.9 0.0X
-nullRatio=0.3, random 14 14
0 75.0 13.3 0.0X
-nullRatio=0.3, clustered 7 7
0 147.9 6.8 0.0X
-nullRatio=0.5, random 16 16
0 67.5 14.8 0.0X
-nullRatio=0.5, clustered 7 7
0 151.9 6.6 0.0X
-nullRatio=0.9, random 9 9
0 121.9 8.2 0.0X
-nullRatio=0.9, clustered 6 6
0 166.4 6.0 0.0X
-nullRatio=1.0, random 0 0
0 4126.2 0.2 0.6X
+nullRatio=0.0, n/a 0 0
0 6438.3 0.2 1.0X
+nullRatio=0.1, random 9 9
0 111.4 9.0 0.0X
+nullRatio=0.1, clustered 6 6
0 164.2 6.1 0.0X
+nullRatio=0.3, random 14 14
0 74.7 13.4 0.0X
+nullRatio=0.3, clustered 6 6
0 164.3 6.1 0.0X
+nullRatio=0.5, random 16 16
0 66.9 14.9 0.0X
+nullRatio=0.5, clustered 6 6
0 172.1 5.8 0.0X
+nullRatio=0.9, random 8 8
0 128.9 7.8 0.0X
+nullRatio=0.9, clustered 5 5
0 194.2 5.1 0.0X
+nullRatio=1.0, random 0 0
0 12165.7 0.1 1.9X
================================================================================================
Nullable batch decode without def-level materialization
================================================================================================
-OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1013-azure
+OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 7763 64-Core Processor
Nullable batch without def-levels: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, n/a 0 0
0 10982.3 0.1 1.0X
-nullRatio=0.1, random 8 8
0 139.7 7.2 0.0X
-nullRatio=0.1, clustered 5 6
0 191.6 5.2 0.0X
-nullRatio=0.3, random 11 11
0 98.9 10.1 0.0X
-nullRatio=0.3, clustered 5 5
0 194.1 5.2 0.0X
-nullRatio=0.5, random 12 12
0 89.4 11.2 0.0X
-nullRatio=0.5, clustered 5 5
0 199.1 5.0 0.0X
-nullRatio=0.9, random 7 7
0 160.5 6.2 0.0X
-nullRatio=0.9, clustered 5 5
0 218.1 4.6 0.0X
-nullRatio=1.0, random 0 0
0 4916.7 0.2 0.4X
+nullRatio=0.0, n/a 0 0
0 11029.7 0.1 1.0X
+nullRatio=0.1, random 7 7
0 159.7 6.3 0.0X
+nullRatio=0.1, clustered 4 4
0 239.6 4.2 0.0X
+nullRatio=0.3, random 10 10
0 107.1 9.3 0.0X
+nullRatio=0.3, clustered 4 4
0 240.9 4.2 0.0X
+nullRatio=0.5, random 11 11
0 95.9 10.4 0.0X
+nullRatio=0.5, clustered 4 4
0 248.0 4.0 0.0X
+nullRatio=0.9, random 6 6
0 186.4 5.4 0.0X
+nullRatio=0.9, clustered 4 4
0 279.2 3.6 0.0X
+nullRatio=1.0, random 0 0
0 21368.1 0.0 1.9X
================================================================================================
Nullable batch decode with row-index filtering (with def-levels)
================================================================================================
-OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1013-azure
+OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 7763 64-Core Processor
Nullable batch with def-levels, row-index filtered: Best Time(ms) Avg
Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
----------------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, contiguous 50% 1
2 0 759.7 1.3 1.0X
-nullRatio=0.3, contiguous 50% 10
10 0 108.5 9.2 0.1X
-nullRatio=0.9, contiguous 50% 7
7 0 149.5 6.7 0.2X
-nullRatio=0.0, alt 1000-row windows 2
3 0 419.6 2.4 0.6X
-nullRatio=0.3, alt 1000-row windows 11
11 0 96.9 10.3 0.1X
-nullRatio=0.9, alt 1000-row windows 8
8 0 128.3 7.8 0.2X
+nullRatio=0.0, contiguous 50% 1
2 0 761.3 1.3 1.0X
+nullRatio=0.3, contiguous 50% 9
9 0 116.5 8.6 0.2X
+nullRatio=0.9, contiguous 50% 6
6 0 171.0 5.8 0.2X
+nullRatio=0.0, alt 1000-row windows 3
3 0 419.3 2.4 0.6X
+nullRatio=0.3, alt 1000-row windows 10
10 0 103.6 9.7 0.1X
+nullRatio=0.9, alt 1000-row windows 7
7 0 144.4 6.9 0.2X
================================================================================================
Nullable batch decode with row-index filtering (without def-levels)
================================================================================================
-OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1013-azure
+OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 7763 64-Core Processor
Nullable batch without def-levels, row-index filtered: Best Time(ms) Avg
Time(ms) Stdev(ms) Rate(M/s) Per Row(ns) Relative
-------------------------------------------------------------------------------------------------------------------------------------
-nullRatio=0.0, contiguous 50% 1
1 0 865.3 1.2 1.0X
-nullRatio=0.3, contiguous 50% 8
8 0 127.6 7.8 0.1X
-nullRatio=0.9, contiguous 50% 6
6 0 174.1 5.7 0.2X
-nullRatio=0.0, alt 1000-row windows 2
2 0 425.3 2.4 0.5X
-nullRatio=0.3, alt 1000-row windows 9
10 0 110.5 9.1 0.1X
-nullRatio=0.9, alt 1000-row windows 7
7 0 143.3 7.0 0.2X
+nullRatio=0.0, contiguous 50% 1
1 0 864.1 1.2 1.0X
+nullRatio=0.3, contiguous 50% 7
7 0 146.8 6.8 0.2X
+nullRatio=0.9, contiguous 50% 5
5 0 208.7 4.8 0.2X
+nullRatio=0.0, alt 1000-row windows 2
3 0 425.0 2.4 0.5X
+nullRatio=0.3, alt 1000-row windows 8
8 0 124.5 8.0 0.1X
+nullRatio=0.9, alt 1000-row windows 6
6 0 167.4 6.0 0.2X
================================================================================================
Single-value reads
================================================================================================
-OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1013-azure
+OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 7763 64-Core Processor
Single-value reads: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-readBoolean 4 4
1 272.0 3.7 1.0X
-readInteger, bitWidth=4 4 4
0 274.0 3.6 1.0X
-readValueDictionaryId, bitWidth=4 4 4
0 275.2 3.6 1.0X
-readInteger, bitWidth=8 4 4
0 272.3 3.7 1.0X
-readValueDictionaryId, bitWidth=8 4 4
0 273.4 3.7 1.0X
-readInteger, bitWidth=12 5 5
0 228.2 4.4 0.8X
-readValueDictionaryId, bitWidth=12 5 5
0 228.9 4.4 0.8X
-readInteger, bitWidth=20 5 5
0 204.3 4.9 0.8X
-readValueDictionaryId, bitWidth=20 5 5
0 205.5 4.9 0.8X
+readBoolean 3 3
0 396.7 2.5 1.0X
+readInteger, bitWidth=4 3 3
0 376.6 2.7 0.9X
+readValueDictionaryId, bitWidth=4 3 3
0 376.6 2.7 0.9X
+readInteger, bitWidth=8 3 3
0 370.6 2.7 0.9X
+readValueDictionaryId, bitWidth=8 3 3
0 370.9 2.7 0.9X
+readInteger, bitWidth=12 4 4
0 292.4 3.4 0.7X
+readValueDictionaryId, bitWidth=12 4 4
1 293.1 3.4 0.7X
+readInteger, bitWidth=20 4 4
0 258.4 3.9 0.7X
+readValueDictionaryId, bitWidth=20 4 4
0 258.1 3.9 0.7X
================================================================================================
Skip
================================================================================================
-OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1013-azure
+OpenJDK 64-Bit Server VM 17.0.19+10-LTS on Linux 6.17.0-1018-azure
AMD EPYC 7763 64-Core Processor
Skip: Best Time(ms) Avg Time(ms)
Stdev(ms) Rate(M/s) Per Row(ns) Relative
------------------------------------------------------------------------------------------------------------------------
-skipBooleans, trueRatio=0.0 0 0
0 21399510.2 0.0 1.0X
-skipBooleans, trueRatio=0.5 2 2
0 536.1 1.9 0.0X
+skipBooleans, trueRatio=0.0 0 0
0 20971520.0 0.0 1.0X
+skipBooleans, trueRatio=0.5 1 1
0 1259.5 0.8 0.0X
skipBooleans, trueRatio=1.0 0 0
0 21399510.2 0.0 1.0X
-skipIntegers PACKED, bitWidth=4 2 2
0 502.5 2.0 0.0X
-skipIntegers RLE, bitWidth=4 0 0
0 20971520.0 0.0 1.0X
-skipIntegers PACKED, bitWidth=8 2 2
0 497.5 2.0 0.0X
+skipIntegers PACKED, bitWidth=4 1 1
0 1070.0 0.9 0.0X
+skipIntegers RLE, bitWidth=4 0 0
0 21399510.2 0.0 1.0X
+skipIntegers PACKED, bitWidth=8 1 1
0 1017.8 1.0 0.0X
skipIntegers RLE, bitWidth=8 0 0
0 21399510.2 0.0 1.0X
-skipIntegers PACKED, bitWidth=12 3 3
0 367.9 2.7 0.0X
+skipIntegers PACKED, bitWidth=12 2 2
0 590.4 1.7 0.0X
skipIntegers RLE, bitWidth=12 0 0
0 17476266.7 0.0 0.8X
-skipIntegers PACKED, bitWidth=20 3 3
0 310.9 3.2 0.0X
+skipIntegers PACKED, bitWidth=20 2 2
0 465.3 2.1 0.0X
skipIntegers RLE, bitWidth=20 0 0
0 17476266.7 0.0 0.8X
diff --git
a/sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/VectorizedRleValuesReader.java
b/sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/VectorizedRleValuesReader.java
index 7c5742b65ada..d1b2726ceeb0 100644
---
a/sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/VectorizedRleValuesReader.java
+++
b/sql/core/src/main/java/org/apache/spark/sql/execution/datasources/parquet/VectorizedRleValuesReader.java
@@ -1000,11 +1000,15 @@ public final class VectorizedRleValuesReader extends
ValuesReader
this.currentBuffer = new int[this.currentCount];
}
currentBufferIdx = 0;
+ // Slice all packed bytes in one call (numGroups groups x bitWidth
bytes each)
+ // instead of one slice per group, avoiding per-group ByteBuffer
allocation.
+ int totalBytes = numGroups * bitWidth;
+ ByteBuffer packed = in.slice(totalBytes);
+ int pos = packed.position();
int valueIndex = 0;
while (valueIndex < this.currentCount) {
- // values are bit packed 8 at a time, so reading bitWidth will
always work
- ByteBuffer buffer = in.slice(bitWidth);
- this.packer.unpack8Values(buffer, buffer.position(),
this.currentBuffer, valueIndex);
+ this.packer.unpack8Values(packed, pos, this.currentBuffer,
valueIndex);
+ pos += bitWidth;
valueIndex += 8;
}
}
diff --git
a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/VectorizedRleValuesReaderSuite.scala
b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/VectorizedRleValuesReaderSuite.scala
index 4ee17fb19c92..5b3997160ece 100644
---
a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/VectorizedRleValuesReaderSuite.scala
+++
b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/VectorizedRleValuesReaderSuite.scala
@@ -20,6 +20,8 @@ package org.apache.spark.sql.execution.datasources.parquet
import java.nio.ByteBuffer
import java.util.PrimitiveIterator
+import scala.jdk.CollectionConverters._
+
import org.apache.parquet.bytes.ByteBufferInputStream
import org.apache.spark.SparkFunSuite
@@ -128,6 +130,61 @@ class VectorizedRleValuesReaderSuite extends SparkFunSuite
{
val page2 = Array.fill(64)(1) ++ Array.tabulate(64)(i => i & 1)
runAndAssertMultiPage(Seq(page1, page2), maxDef = 1, batchSize = 64)
}
+
+ test("PACKED: multi-buffer stream with packed run crossing buffer boundary")
{
+ // Exercises the MultiBufferInputStream.slice() path where the packed
bytes span a buffer
+ // boundary, causing slice() to return a freshly-allocated contiguous
buffer with
+ // position() == 0 (the base-0 pos branch in readNextGroup).
+ val n = 256
+ val defLevels = Array.tabulate(n)(i => i & 1)
+ val bitWidth = 1
+ val encoded = encodeRle(defLevels, bitWidth)
+
+ // Split the encoded bytes into small chunks (e.g. 3 bytes each) so the
packed data
+ // is guaranteed to span at least one buffer boundary within
MultiBufferInputStream.
+ val chunkSize = 3
+ val buffers = encoded.grouped(chunkSize).map { chunk =>
+ ByteBuffer.wrap(chunk)
+ }.toList.asJava
+
+ val nonNullCount = defLevels.count(_ == 1)
+ val plainBytes = plainIntBytes(nonNullCount)(valueAt)
+
+ val reader = new VectorizedRleValuesReader(bitWidth, false)
+ reader.initFromPage(n, ByteBufferInputStream.wrap(buffers))
+ val valueReader = new VectorizedPlainValuesReader
+ valueReader.initFromPage(
+ nonNullCount, ByteBufferInputStream.wrap(ByteBuffer.wrap(plainBytes)))
+ val state = ParquetTestAccess.newState(intColumnDescriptor(1), false)
+ ParquetTestAccess.resetForNewPage(state, n, 0L)
+
+ val batchSize = 64
+ var produced = 0
+ var expectedValueIdx = 0
+ while (produced < n) {
+ val toRead = math.min(batchSize, n - produced)
+ val values = new OnHeapColumnVector(toRead, IntegerType)
+ ParquetTestAccess.resetForNewBatch(state, toRead)
+ ParquetTestAccess.readBatch(reader, state, values, null, valueReader,
integerUpdater)
+
+ var i = 0
+ while (i < toRead) {
+ val absPos = produced + i
+ if (defLevels(absPos) == 1) {
+ assert(!values.isNullAt(i), s"pos $absPos should be non-null")
+ val expected = valueAt(expectedValueIdx)
+ assert(
+ values.getInt(i) == expected,
+ s"pos $absPos value mismatch: got ${values.getInt(i)}, expected
$expected")
+ expectedValueIdx += 1
+ } else {
+ assert(values.isNullAt(i), s"pos $absPos should be null")
+ }
+ i += 1
+ }
+ produced += toRead
+ }
+ }
}
private object VectorizedRleValuesReaderSuite {
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
