The logic behind bulk insert optimization is simple.
Instead of writing each key value to B-tree (that is, to the key cache, although the bulk insert code doesn't know about the key cache), we store keys in a balanced binary (red-black) tree, in memory. When this tree reaches its memory limit, we write all keys to disk (to key cache, that is). But since the key stream coming from the binary tree is already sorted, inserting goes much faster, all the necessary pages are already in cache, disk access is minimized, and so forth.
Copyright © 1997, 2014, Oracle and/or its affiliates. All rights reserved. Legal Notices