I'm working on implementing a few optimizations for our DWH, in particular I want to automatically find the best partition and bucketing schema for tables based on their access patterns. This is actually part of a much bigger project that I want to do to further increase the performance and efficiency of our data analytics systems.