I am following book 'Definitive Cassandra'. A hotel application is used as an example in it. THere is a table available_rooms_by_hotel_date. It is to support use case when user wants to know about room availability in a given hotel from a given date. Data model is defined as:
hotel_id
date
room_number
is_available
hotel_id is partition key, while date and room_number are clustering columns.
Looking at the table, one can say that it supports use case when user wants to know availability of room from a given date for a given hotel.
I understand that order of clustering columns is also critical in Cassandra. So if I change order for date and room_number, how would it impact? Functionality wise I think use case is still supported. But does it impact performance or any other aspects like storage, node allocation etc?