with most people working from home, not sure if this heuristic works.

disclaimer: works for databricks, but not on spark, and first time posting in this thread