The Percentiles function creates percentiles for groups of numbers. The nth percentile is the smallest value in a data set that is greater than n% of the values.
Use this function when the number of input data partitions is large and you want to find the percentile for each group. Each group must fit on a single worker node. The maximum number of input rows in each group that the function can process depends on the cluster configuration. To find percentile statistics for a very large input group that requires multiple workers, use the function Approximate Percentile.