Google Cloud Serverless for Apache Spark documentation
Use Serverless for Apache Spark to run Spark batch workloads without provisioning and managing your own cluster. Specify workload parameters, and then submit the workload to the Serverless for Apache Spark service. The service will run the workload on a managed compute infrastructure, autoscaling resources as needed. Serverless for Apache Spark charges apply only to the time when the workload is executing. Learn more
Start your proof of concept with $300 in free credit
- Get access to Gemini 2.0 Flash Thinking
- Free monthly usage of popular products, including AI APIs and BigQuery
- No automatic charges, no commitment
Keep exploring with 20+ always-free products
Access 20+ free products for common use cases, including AI APIs, VMs, data warehouses, and more.
Documentation resources
Related resources
Serverless for Apache Spark Quickstart Interactive Tutorial
Interactive tutorial for getting started with Serverless for Apache Spark.
Dataproc Templates on GitHub
Serverless for Apache Spark ready-to-use, config-driven Spark templates.
Serverless for Apache Spark Labs on GitHub
Serverless for Apache Spark hands-on labs built around common use cases.