Is it possible to cache data in Spark?

Question

Java

For the REST API, I used the Spark framework.
Each client request creates a request for a more complex data source that doesn't "spit out" data very quickly.
Therefore, when there are many requests, the responses come very slowly.
Is there any way to cache the data?

Answer 1

hazelcast, redis, tarantool, etc

Answer 2

In the book of Holden Karau - "Learning Spark" - chapter 9 of Spark SQL, 1-2 pages are devoted to this issue. You can read.
You can cache "somehow" "anything". Redis and other boxed solutions for caches work,
but they have nothing to do with Spark technology. Actually, they can be advised as a universal
advice for any systems that have a back-end. Their integration with Spark is generally a separate issue.