Storing a Trie with Compact and Predictable Space

TL;DR

This paper introduces a coordinate hash trie, a space-efficient and predictable storage method for trie structures that achieves constant average-time operations and is simple to implement.

Contribution

It presents a novel trie storage approach using a global hash table with predictable space and constant average operation time, improving over traditional methods.

Findings

Space complexity is O(n), independent of alphabet size m.

Average case operation time is O(1).

Approach is simple to implement.

Abstract

This paper proposed a storing approach for trie structures, called coordinate hash trie. The basic idea is using a global hash table with a special hash function to store all edges of a trie. For a trie with $n$ nodes and an alphabet with size $m$, the execution time of finding, inserting and deleting a child node, is $O(1)$ for the average case, $O(m)$ for the worst case. The space used by this approach is $O(n)$, unrelated to $m$. The constant of space consumption is predictable, with no need for reallocation or resizing. In addition, this approach is very easy to implement.