Parallel implementation of the trie structure

We present two simple algorithms for implementing the trie structure based on the MIMD (Multiple Instruction Streams Multiple Data Streams) model of computation. Unlike the implementation, our schemes allow operations on the trie structure by multiple processes. No hash function will be employed and the problem due to collision is thus, eliminated. By making use of a simple segmentation scheme. We are able to garbage collect and recycle unused memory space in an efficient manner. Results from simulation show that our proposed algorithms attain satisfactory speedup.