High Performance Deflate Decompression on Intel Architecture Processors

Date Added: Nov 2010
Format: PDF

There is a critical need for lossless data compression in enterprise storage and applications such as databases which process huge amounts of data. DEFLATE is a widely used standard to perform lossless compression and forms the basis of utilities such as gzip and libraries such as zlib. In these applications, decompression imposes a large computational burden on the servers and they could benefit from a highly optimized implementation. This paper describes the performance characteristics of an optimized implementation of DEFLATE decompression, on Intel processors based on the 32 nm micro architecture. As the performance of decompression is data dependent, the authors report the performance on various industry standard corpora data sets.