Undersantanding Redis Cache Speed and Priority

Hi guys, sorry if this is not a good question, but I’m trying to understand a behavior that has been happening.
Just a disclaimer, I’m not involved in Redis administration in my company, but my application consumes a lot from Redis.

I have a bunch of servers querying Redis and for simplification the query consists of a pair of keys (key1,key2) and I have a limited number of combinations:
(01, 01)
(01, 02)
(02, 03)
(02, 04)
(03, 05)
(03, 06)

but they are not requested at the same proportion, so 90% of my requests do Redis will be query (01,01).
Will the query that is most used 90% of the time be faster than a query I have to use almost never?

Redis stores all of it’s keys and values in a giant hash table. Assuming these keys (or key pairs) contain comparable data, I would expect the performance to be the same regardless of which set of keys are being retrieved.

Test the pairs of keys off the equation, find the negative and replace with a positive.