In computing theory, some techniques, such as LRU (Least Recently Used), MRU (Most Recently Used), are quite common when it comes to improving performance using the concept of caching. The idea is to temporarily persist the frequently used data at the client site. In turn, the system is not expected to go and retrieve the data on every transaction. Instead, it can refer the cached data in the local layer.

What is the benefit of doing this? Eventually, it improves overall system performance by saving a large amount of computing time and network latency.