Importing Microsoft SQL Server tables with datetime columns causes high cpu and memory usage.

When importing a sql server table with datetime columns the cpu and memory usage spikes. This causes power bi and Windows in general to become unresponsive. The more datetime columns in the table the higher the memory spikes and the cpu is maxed for a greater length of time. For example, my data set has a table with 26 datetime columns, in addition to other column types, and power bi desktop returns an out of memory error when importing this table. If the query is limited to exclude the datetime columns then the table is imported quickly with little additional cpu and memory usage. As the datetime columns are included back into the query, cpu usage and memory usage increase at a near linear rate until the out of memory error occurs. This happens at approx. 20 datetime type columns.