In the Understanding Elasticsearch caching section of Chapter 6, Low-level Index Control we described caching: one of many ways that allow us to improve Elasticsearch's outstanding performance. Unfortunately, caching is not a silver bullet and, sometimes, it is better to avoid it. If your data is changing rapidly and your queries are very unique and not repeatable, then caching won't really help and can even make your performance worse sometimes.
Every cache is based on a simple principle. The main assumption is that to improve performance, it is worth storing some part of the data in the memory instead of fetching from slow sources such as spinning disks, or to save the system a need to recalculate some processed data. However, caching is not free and it has its price—in terms of Elasticsearch, the cost of caching is mostly memory. Depending on the cache type, you may only need to store recently used data, but again...