Spark job scales linearly during Sqoop import and export. You can tune Spark jobs based on cluster resources. Configure the following advanced properties for Spark in the Hadoop connection:

spark.executor.instances=<number of executor instances>

The following formula determines the total running containers:

Total running containers = (Number of cores) x (Number of executor instances)

The Spark engine uses 2 executor instances by default. So, only 4 containers run in parallel. For better performance, fine tune the spark.executor.instances property.