How should we handle caching for data that's read a lot, but doesn't change much?

hey folks, i'm working on a system that has a lot of read operations on relatively static data. we're seeing some performance bottlenecks during peak times. i've been thinking about implementing a caching layer, but i'm not sure about the best approach. should i go with in-memory caching like Redis or Memcached, or something more distributed? what are your go-to strategies for managing cache invalidation with read-heavy workloads? any insights on trade-offs would be super helpful