Implementation and Evaluation of String B-Tree

String B-tree is a combination of B-tree and Patricia tries for internal-node indices. Instead of storing prefix compressed keys at each index node, each key is stored in full in a consecutive sequence of data blocks, and each downward-traversal decision is made by a combination of Patricia trie search and the consultation of a single key. String B-tree has the same worst case performance as B-tree but it manages unbounded-length strings and performs much more powerful search operations such as the ones supported by suffix tree. In this project, we implemented a static string B-tree for string search in external memory. The operations supported include Prefix Search, Range Query, and Substring Search. Our experiment compared the query performance of String B-tree with B+-Tree and linear search. The result shows String B-tree outperforms B+-tree by reducing disk access admirably.