I have implemented TST (ternary search tree) in C to create fast dictionary (histogram) for counting words in Wikipedia. It is fast and memory efficient.