Low-Complexity Lossless Compression on High-Speed Networks

We present a survey of results on how to implement low-complexity lossless data compression on a high speed network, so that the computational phase requires no interprocessor communication. It follows that the computation in between the input and output phases has a linear speed-up when the network size increases, regardless of the bandwidth and latency of the network. Depending on the type of data, the performance of the compression method changes in terms of scalability. Images are more suitable than strings, since text compression is scalable only on very large size files. Keywords-high speed network application; lossless compression; distributed algorithm; scalability.