Skip to content

Enable precision 4 for HLLPP #14430

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
res-life merged 3 commits into from
Mar 31, 2026
Merged

Enable precision 4 for HLLPP #14430

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
c3469ee
Fix HLLPP bug when precision is 4
Mar 17, 2026
a1ed823
Update sql-plugin/src/main/scala/com/nvidia/spark/rapids/GpuOverrides…
res-life Mar 18, 2026
7ac5cca
Format
Mar 31, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 3 additions & 7 deletions integration_tests/src/main/python/hyper_log_log_plus_plus_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
# Copyright (c) 2025, NVIDIA CORPORATION.
# Copyright (c) 2025-2026, NVIDIA CORPORATION.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
Expand Down Expand Up @@ -40,7 +40,7 @@ def test_hllpp_reduction(data_gen):

# precision = Math.ceil(2.0d * Math.log(1.106d / relativeSD) / Math.log(2.0d)).toInt
_relativeSD = [
pytest.param(0.3, marks=pytest.mark.xfail(reason="https://github.com/NVIDIA/spark-rapids/issues/12452")), # precision 4
0.3, # precision 4
0.25, # precision 5
0.15, # precision 6
0.1, # precision 7
Expand Down Expand Up @@ -68,11 +68,7 @@ def test_hllpp_precisions_reduce(relativeSD):

@pytest.mark.skipif(is_databricks_runtime(), reason="HyperLogLogPlusPlus does not support Databricks currently(https://github.com/NVIDIA/spark-rapids/issues/12388)")
@ignore_order(local=True)
@pytest.mark.parametrize('relativeSD',
[x if x != 0.3
else pytest.param(x, marks=pytest.mark.xfail(reason="https://github.com/NVIDIA/spark-rapids/issues/12452"))
for x in _relativeSD],
ids=idfn)
@pytest.mark.parametrize('relativeSD', _relativeSD, ids=idfn)
def test_hllpp_precisions_groupby(relativeSD):
assert_gpu_and_cpu_are_equal_sql(
lambda spark: gen_df(spark, [("c1", int_gen), ("c2", int_gen)]),
Expand Down
14 changes: 6 additions & 8 deletions sql-plugin/src/main/scala/com/nvidia/spark/rapids/GpuOverrides.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4125,14 +4125,12 @@ object GpuOverrides extends Logging {
}
val precision = GpuHyperLogLogPlusPlus.computePrecision(a.relativeSD)
// Spark supports precision range: [4, Infinity)
// Spark-Rapids only supports precision range: [5, 14]
if (precision <= 4 || precision > 14) {
//
// Info: cuCollection supports precision range [4, 18]
// Due to https://github.com/NVIDIA/spark-rapids/issues/12347, the Spark-Rapids supports
// fewer precisions than cuCollection: range: [5, 14]
willNotWorkOnGpu(s"The precision $precision from relativeSD ${a.relativeSD} is bigger" +
s" than 14, GPU only supports precision range [5, 14].")
// cuCollection supports precision range: [4, 18]
// Spark-Rapids only supports precision range: [4, 14],
// GPU does not perform well for precision > 14.
if (precision < 4 || precision > 14) {
willNotWorkOnGpu(s"The precision $precision from relativeSD ${a.relativeSD} is out of" +
s" range, GPU only supports precision range [4, 14].")
}
}

Expand Down
Loading