Does the Redis package make use of connection pooling when used for caching in CF (2023)? Curious if, for example, iterating over an array of IDs and calling cacheGet() each time is leveraging the opportunity for fewer connections by referencing from a connection pool. I ask mostly to have a better idea of performance considerations when using CF's function versus dropping to Jedis directly, creating a pool instance, and then asking for a resource to run a series of gets when needed.