Parallel Data Compression

Parallel computing allows software to run faster by using multiple processors and performing multiple computations simultaneously when possible. The majority of general purpose software is designed for sequential machines and does not take advantage of parallel computing. Lossless data compression software like WinZip, gzip, and bzip2, used for compressing any files found on a computer, are designed for sequential machines. The data compression algorithms require a lot of processing power to analyze and then encode the files in a smaller form. Writing a general purpose compressor that can take advantage of parallel computing should greatly reduce the amount of time it requires to compress files, especially large ones. A parallel version of bzip2 shall be created and tested on a dual-processor machine and parallel cluster to determine if and how much speedup is achieved. The bzip2 program was chosen because it achieves very good compression and is available in library and source form, free under a BSD-style license. The limited research papers available on parallel data compression use theoretical algorithms designed to run on impractical parallel machines that do not exist or are very rare in the real world. The idea with the parallel bzip2 compression program is to make a practical parallel compression utility that works with real machines.