Skip to content
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

Jump to bottom

[VL][CI] Fix fallback to columnar shuffle in celeborn test #4474

Merged
merged 2 commits into from
Jan 29, 2024
Merged

[VL][CI] Fix fallback to columnar shuffle in celeborn test #4474

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
bfe324b
Initial commit
PHILO-HE Jan 22, 2024
ff1a297
Test reverting the change for VeloxColumnarBatch.cc
PHILO-HE Jan 29, 2024
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: 6 additions & 4 deletions .github/workflows/velox_be.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -372,10 +372,12 @@ jobs:
- name: TPC-H SF1.0 && TPC-DS SF10.0 Parquet local spark3.2 with Celeborn
run: |
$PATH_TO_GLUTEN_TE/$OS_IMAGE_NAME/gha/gha-checkout/exec.sh \
'mv /opt/apache-celeborn-0.3.0-incubating-bin/conf/celeborn-env.sh.template /opt/apache-celeborn-0.3.0-incubating-bin/conf/celeborn-env.sh && \
echo -e "CELEBORN_MASTER_MEMORY=4g\nCELEBORN_WORKER_MEMORY=4g\nCELEBORN_WORKER_OFFHEAP_MEMORY=8g" > /opt/apache-celeborn-0.3.0-incubating-bin/conf/celeborn-env.sh && \
echo -e "celeborn.worker.commitFiles.threads 128\nceleborn.worker.sortPartition.threads 64" > /opt/apache-celeborn-0.3.0-incubating-bin/conf/celeborn-defaults.conf \
&& bash /opt/apache-celeborn-0.3.0-incubating-bin/sbin/start-master.sh && bash /opt/apache-celeborn-0.3.0-incubating-bin/sbin/start-worker.sh && \
'wget https://dlcdn.apache.org/incubator/celeborn/celeborn-0.3.0-incubating/apache-celeborn-0.3.0-incubating-bin.tgz && \
tar xzf apache-celeborn-0.3.0-incubating-bin.tgz && cd apache-celeborn-0.3.0-incubating-bin && \
mv ./conf/celeborn-env.sh.template ./conf/celeborn-env.sh && \
echo -e "CELEBORN_MASTER_MEMORY=4g\nCELEBORN_WORKER_MEMORY=4g\nCELEBORN_WORKER_OFFHEAP_MEMORY=8g" > ./conf/celeborn-env.sh && \
echo -e "celeborn.worker.commitFiles.threads 128\nceleborn.worker.sortPartition.threads 64" > ./conf/celeborn-defaults.conf \
&& bash ./sbin/start-master.sh && bash ./sbin/start-worker.sh && \
cd /opt/gluten/tools/gluten-it && mvn clean install -Pspark-3.2,rss \
&& GLUTEN_IT_JVM_ARGS=-Xmx5G sbin/gluten-it.sh queries-compare \
--local --preset=velox-with-celeborn --benchmark-type=h --error-on-memleak --off-heap-size=10g -s=1.0 --threads=16 --iterations=1 \
Expand Down
23 changes: 12 additions & 11 deletions cpp/velox/memory/VeloxColumnarBatch.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,6 @@
*/
#include "VeloxColumnarBatch.h"
#include "compute/VeloxRuntime.h"
#include "utils/Timer.h"
#include "utils/VeloxArrowUtils.h"
#include "velox/row/UnsafeRowFast.h"
#include "velox/type/Type.h"
Expand Down Expand Up @@ -48,17 +47,20 @@ void VeloxColumnarBatch::ensureFlattened() {
if (flattened_ != nullptr) {
return;
}

ScopedTimer timer(&exportNanos_);
auto startTime = std::chrono::steady_clock::now();
// Make sure to load lazy vector if not loaded already.
for (auto& child : rowVector_->children()) {
facebook::velox::BaseVector::flattenVector(child);
if (child->isLazy()) {
child = child->as<facebook::velox::LazyVector>()->loadedVectorShared();
VELOX_DCHECK_NOT_NULL(child);
}
child->loadedVector();
}
// Invalid the original rowVector_ after being flattened.
flattened_ = std::move(rowVector_);

// Perform copy to flatten dictionary vectors.
velox::RowVectorPtr copy = std::dynamic_pointer_cast<velox::RowVector>(
velox::BaseVector::create(rowVector_->type(), rowVector_->size(), rowVector_->pool()));
copy->copy(rowVector_.get(), 0, 0, rowVector_->size());
flattened_ = copy;
auto endTime = std::chrono::steady_clock::now();
auto duration = std::chrono::duration_cast<std::chrono::nanoseconds>(endTime - startTime).count();
exportNanos_ += duration;
}

std::shared_ptr<ArrowSchema> VeloxColumnarBatch::exportArrowSchema() {
Expand All @@ -81,7 +83,6 @@ int64_t VeloxColumnarBatch::numBytes() {
}

velox::RowVectorPtr VeloxColumnarBatch::getRowVector() const {
VELOX_CHECK_NOT_NULL(rowVector_);
return rowVector_;
}

Expand Down
2 changes: 0 additions & 2 deletions tools/gluten-te/ubuntu/dockerfile-buildenv
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -115,8 +115,6 @@ RUN cd /opt/spark331 && ./build/mvn -Pyarn -DskipTests clean install
RUN cd /opt && git clone --depth 1 --branch v3.4.2 https://github.com/apache/spark.git spark342
RUN cd /opt/spark342 && ./build/mvn -Pyarn -DskipTests clean install

RUN cd /opt && git clone --depth 1 --branch v0.3.0-incubating https://github.com/apache/incubator-celeborn.git apache-celeborn-0.3.0-incubating-bin

# Prepare entry command
COPY scripts/cmd.sh /root/.cmd.sh
CMD ["/root/.cmd.sh"]
Loading