Run the queries in parallel

[nflucian]

By running concurrent jobs with a single spark session, will not only maximise the resource utilisation but also reduce application time and cost drastically. Furthermore, if we have adequate resources and these jobs do not have any interlink between them then it does not make sense to execute them in a loop or as a different spark applications.

https://medium.com/analytics-vidhya/boosting-apache-spark-application-by-running-multiple-parallel-jobs-25d13ee7d2a6

[nflucian]

https://www.slideshare.net/databricks/parallelizing-with-apache-spark-in-unexpected-ways