Java – how to store the inverted document index on disk?

I know this question has been asked again and again in stack overflow and Google, but I find that all the answers can't satisfy me Most solutions assume that the entire index can fit into memory, and then we can store it to disk through Java serialization When an index is needed, we must load the entire index into memory Such a solution:

I would appreciate it if you could use java to solve my solution

Solution

I will try jdbm 3, which supports trees and hash sets. The only requirement is that each key or entry is suitable for memory

If you have large entries, I recommend that you store each entry as a memory mapped file to extract some data In the lookup table, you can store the key in a file name (or name the file key)