You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm currently thinking about using a Java based ART implementation as an in-memory index for my Open Source temporal data store (https://sirix.io), which uses an ART like trie for storing paged variable sized document index on flash drives. I'm however also using Bitmap nodes as in hash array mapped tries or Judy arrays.
Now, besides a super cache unfriendly AVL tree as a secondary in memory data structure I'm considering using ART or HOT. However, I don't understand why performance for 95% range scans and 5% inserts is worse in comparison with a TreeMap, which is a red/black binary tree implementation:
Hi,
I'm currently thinking about using a Java based ART implementation as an in-memory index for my Open Source temporal data store (https://sirix.io), which uses an ART like trie for storing paged variable sized document index on flash drives. I'm however also using Bitmap nodes as in hash array mapped tries or Judy arrays.
Now, besides a super cache unfriendly AVL tree as a secondary in memory data structure I'm considering using ART or HOT. However, I don't understand why performance for 95% range scans and 5% inserts is worse in comparison with a TreeMap, which is a red/black binary tree implementation:
https://github.com/rohansuri/adaptive-radix-tree
You'll find the performance comparison diagram when you scroll down. Maybe you have an idea...
The text was updated successfully, but these errors were encountered: