I need a KQL query returning top X records each grouping

Assume a table called Cities like the following:

GA   Augusta       203953
GA   Athens        130873
GA   Atlanta       522328
GA   Savannah      151227
FL   Orlando       328354
FL   Tampa         399734
FL   Miami         455075
FL   Jacksonville  987960

I need a KQL query returning the top 2 cities ranked by population for each state.
For example, here is the expected result:

FL   Jacksonville  987960
FL   Miami         455075
GA   Atlanta       522328
GA   Augusta       203953

>Solution :

you can use the partition operator

for example:

datatable(state:string, city:string, population:long)
[
    "GA", "Augusta", 203953,
    "GA", "Athens", 130873,
    "GA", "Atlanta", 522328,
    "GA", "Savannah", 151227,
    "FL", "Orlando", 328354,
    "FL", "Tampa", 399734,
    "FL", "Miami", 455075,
    "FL", "Jacksonville", 987960,
]
| partition by state ( top 2 by population desc )