I'm studying B+trees and I'm trying to understand how actual data can be stored in a physical file and still allow fast lookups. None of it would be in memory, all of the pointers would be "seeks" to different portions of the file.

Specifically I'm trying to figure out a way to do text searches. Hence the leaf nodes would be an inverted index:

enter image description here

Because I'm storing this all in a single file I'm having problems understanding where in the file the actual tree should go (the pointers) and where the actual data should go. Should the tree simply be the first part of the file and then just have the data? Are there standards for this?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.