RAM is expensive and fast. Hard disk drives or solid state drives are relatively cheaper and slower and also provide durability of our data in the case of system and power failure. All our data is stored on disk and when we perform a query, MongoDB will try and fetch data from memory. If data is not in memory then it will fetch data from disk and copy it to memory. This is a page fault event, because of the fact that data in memory is organized in pages.

As page faults happen, memory gets filled up and eventually some pages need to be cleared for more recent data to come into memory. This is called a page eviction event. We cannot completely avoid page faults unless we have a really static dataset but we want to try and minimize page faults. This can be achieved by holding our working set in memory.