CUDA-Compression

As the name suggests, CUDA-Compressions is a GPU-based compression algorithm, which has been highly tuned for NVIDIA GPUs and load sharing between CPUs and GPUs. The compression kernel can be used in a software pipeline to compress data before it is sent across a network or to another I/O device. The decompression kernel can analogously be used to receive compressed data.

The initial code for this work was developed at Delaware with Prof. Martin Swany's group. However, it has been greatly enhanced since then in the last year at IU. A large majority of the current code is the result of these enhancements. The software consists of more than 13,000 lines of C and CUDA code, which includes certain benchmarks.

A paper from this effort was published at ICPADS, and was among the top papers from the conference to be invited for journal submission.

Related publications:

1. Adnan Ozsoy and Martin Swany and Arun Chauhan.. Optimizing LZSS Compression on GPGPUs. Future Generation Computer Systems (FGCS), 2013.
2. Adnan Ozsoy, Martin Swany and Arun Chauhan. Pipelined Parallel LZSS for Streaming Data Compression on GPGPUs. In Proceedings of the 18h IEEE International Conference on Parallel and Distributed Systems (ICPADS), 2012.
   [Aricle DOI]
   Paper invited for submission as a journal article.