Modern lossless compression techniques: Review, comparison and analysis

With the world drifting more and more towards the social network, the size and amount of data shared over the internet is increasing day by day. Since the network bandwidth is always limited, we require efficient compression algorithms to facilitate fast and efficient sharing of data over the network. In this paper, we discuss algorithms of widely used traditional and modern compression techniques. In an effort to find the optimum compression algorithm, we compare commonly used modern compression algorithms: Deflate, Bzip2, LZMA, PPMd and PPMonstr by analyzing their performance on Silesia corpus. The corpus comprises of files of varied type and sizes, which accurately simulates the vast diversity of files shared over the internet. We compare these algorithms on the basis of their compression ratio, compression speed and decompression speed. After observing the simulated results, we found that PPMonstr provides the best compression ratio. Deflate is the fastest algorithm in terms of compression and decompression speed, but provides low compression ratio. Bzip2 and PPMd provide moderate compression speed and good compression ratio and hence are well suited for applications dependent on both compression ratio and speed. Further, experimental results indicate that LZMA is an ideal algorithm for applications requiring high compression ratio along with fast decompression speed as it provides moderate decompression speed coupled with high compression ratio.