Spark in java that list files in GCS modify the parameters in run.sh accordingly to your project make build to build a monolithc jar make run_serverless to run this job in Dataproc serverless mode