Skip to content

[GLUTEN-10660][VL] Adding configuration for hash table build #10634

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
zhouyuan merged 6 commits into from
Sep 29, 2025
Merged

[GLUTEN-10660][VL] Adding configuration for hash table build #10634

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
a219f72
[VL] Adding configuration for hash table build
zhouyuan Aug 29, 2025
81a9d6c
fix default value
zhouyuan Sep 8, 2025
f030c73
set as experimental
zhouyuan Sep 18, 2025
7b0a64a
update docs
zhouyuan Sep 18, 2025
0b3bee5
disable by default
zhouyuan Sep 24, 2025
de42175
fix default value
zhouyuan Sep 29, 2025
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
14 changes: 14 additions & 0 deletions backends-velox/src/main/scala/org/apache/gluten/config/VeloxConfig.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -529,6 +529,20 @@ object VeloxConfig {
.booleanConf
.createWithDefault(false)

val VELOX_HASHMAP_ABANDON_BUILD_DUPHASH_MIN_ROWS =
buildConf("spark.gluten.velox.abandonbuild.noduphashminrows")
.experimental()
.doc("Experimental: abandon hashmap build if duplicated rows more than this number.")
.intConf
.createWithDefault(100000)

val VELOX_HASHMAP_ABANDON_BUILD_DUPHASH_MIN_PCT =
buildConf("spark.gluten.velox.abandonbuild.noduphashminpct")
.experimental()
.doc("Experimental: abandon hashmap build if duplicated rows are more than this percentile.")
.doubleConf
.createWithDefault(0)

val QUERY_TRACE_ENABLED = buildConf("spark.gluten.sql.columnar.backend.velox.queryTraceEnabled")
.doc("Enable query tracing flag.")
.booleanConf
Expand Down
6 changes: 6 additions & 0 deletions cpp/velox/compute/WholeStageResultIterator.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -593,6 +593,12 @@ std::unordered_map<std::string, std::string> WholeStageResultIterator::getQueryC
configs[velox::core::QueryConfig::kMaxSplitPreloadPerDriver] =
std::to_string(veloxCfg_->get<int32_t>(kVeloxSplitPreloadPerDriver, 2));

// hashtable build optimizations
configs[velox::core::QueryConfig::kAbandonBuildNoDupHashMinRows] =
std::to_string(veloxCfg_->get<int32_t>(kAbandonBuildNoDupHashMinRows, 100000));
configs[velox::core::QueryConfig::kAbandonBuildNoDupHashMinPct] =
std::to_string(veloxCfg_->get<int32_t>(kAbandonBuildNoDupHashMinPct, 0));

// Disable driver cpu time slicing.
configs[velox::core::QueryConfig::kDriverCpuTimeSliceLimitMs] = "0";

Expand Down
4 changes: 4 additions & 0 deletions cpp/velox/config/VeloxConfig.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -60,6 +60,10 @@ const std::string kAbandonPartialAggregationMinPct =
const std::string kAbandonPartialAggregationMinRows =
"spark.gluten.sql.columnar.backend.velox.abandonPartialAggregationMinRows";

// hashmap build
const std::string kAbandonBuildNoDupHashMinRows = "spark.gluten.velox.abandonbuild.noduphashminrows";
const std::string kAbandonBuildNoDupHashMinPct = "spark.gluten.velox.abandonbuild.noduphashminpct";

// execution
const std::string kBloomFilterExpectedNumItems = "spark.gluten.sql.columnar.backend.velox.bloomFilter.expectedNumItems";
const std::string kBloomFilterNumBits = "spark.gluten.sql.columnar.backend.velox.bloomFilter.numBits";
Expand Down
Loading