Skip to content

Add dataproc tpcds example notebook #607

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

Open
viadea wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Add dataproc tpcds example notebook #607

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
85db578
Add dataproc tpcds example notebook
Nov 19, 2025
7eac061
Cleared the notebook output and did some minor change on README.
Nov 20, 2025
6d9aa88
Modify some format issue for README
Nov 20, 2025
19d0c6e
Clear a cell output
Nov 20, 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
59 changes: 59 additions & 0 deletions examples/SQL+DF-Examples/tpcds/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,3 +26,62 @@ Here is the bar chart from a recent execution on Google Colab's T4 High RAM inst
RAPIDS Spark 25.10.0 with Apache Spark 3.5.0

![tpcds-speedup](/docs/img/guides/tpcds.png)

## Execute the notebook on Dataproc

### 1. Create a Dataproc cluster

```
export OS_VERSION=ubuntu22
export CLUSTER_NAME=test-$OS_VERSION
export GCS_BUCKET=mybucket
export REGION=us-central1
export ZONE=us-central1-a
export NUM_GPUS=1
export NUM_WORKERS=2

PROPERTIES=(
"spark:spark.history.fs.logDirectory=gs://$GCS_BUCKET/eventlog/"
"spark:spark.eventLog.dir=gs://$GCS_BUCKET/eventlog/"
"spark:spark.history.fs.gs.outputstream.type=FLUSHABLE_COMPOSITE"
"spark:spark.history.fs.gs.outputstream.sync.min.interval.ms=60000"
"spark:spark.driver.memory=20g"
"spark:spark.executor.memory=42g"
"spark:spark.executor.memoryOverhead=8g"
"spark:spark.executor.cores=16"
"spark:spark.executor.instances=2"
"spark:spark.task.resource.gpu.amount=0.001"
"spark:spark.sql.files.maxPartitionBytes=512M"
"spark:spark.rapids.memory.pinnedPool.size=4g"
"spark:spark.shuffle.manager=com.nvidia.spark.rapids.spark353.RapidsShuffleManager"
"spark:spark.jars.packages=ch.cern.sparkmeasure:spark-measure_2.12:0.27"
)

gcloud dataproc clusters create $CLUSTER_NAME \
--region $REGION \
--zone $ZONE \
--image-version=2.3-$OS_VERSION \
--master-machine-type n1-standard-16 \
--master-boot-disk-size 200 \
--num-workers $NUM_WORKERS \
--worker-accelerator type=nvidia-tesla-t4,count=$NUM_GPUS \
--worker-machine-type n1-standard-16 \
--num-worker-local-ssds 2 \
--initialization-actions gs://goog-dataproc-initialization-actions-${REGION}/spark-rapids/spark-rapids.sh \
--optional-components=JUPYTER,ZEPPELIN \
--metadata gpu-driver-provider="NVIDIA",rapids-runtime="SPARK" \
--no-shielded-secure-boot \
--bucket $GCS_BUCKET \
--subnet=default \
--properties="$(IFS=,; echo "${PROPERTIES[*]}")" \
--enable-component-gateway

```

Note: Please adjust the value of `spark.shuffle.manager` based on the Spark version of the Dataproc cluster version.

### 2. Execute the example notebook in Jupyter lab

[TPCDS-SF3K-Dataproc.ipynb](notebooks/TPCDS-SF3K-Dataproc.ipynb)


Loading