Databricks has an extremely bad API. So, sure, your Spark jobs might be a little bit faster some times, but why would you use it if you can't even read logs of

Databricks has an extremely bad API. So, sure, your Spark jobs might be a little bit faster some times, but why would you use it if you can't even read logs of running jobs?