I've been reading about SQL Server internals and every book or blog mentions this about backward scans.
A backward scan of a clustered index cannot use parallelism
The only post that said something is this one below. The post says that the SQL Server team did not implement the required optimizations for a backward scan. https://www.itprotoday.com/sql-server/descending-indexes
Since the leaf level pages are linked using a doubly linked list, I do not understand why a backward scan is different to a forward scan. Any clarification is really appreciated.