Currently the distinct() command finds the unique set of values for a SINGLE specified field across a collection or view. For example: db.staff.distinct("last_name" ) If there is an index on the last_name field, the DISTINCT_SCAN plan can use that index and the operation is very fast. To find the unique values for a set of more than one fields, the $group aggregation stage has to be used like this: db.staff.aggregate([ {$group: {_id: {FName: "$first_name", LName: "$last_name"}} ]); This operation does not really need the $group functionality, as it is not calculating a sum/min/max/average/etc value using the accumulator operators. It also runs much more slowly than a DISTICT_SCAN. The ask here is to allow distinct() to be run over a set of fields, and use DISTINCT_SCAN if there is an index over the same fields.