An implementation of Lz77 compression algorithm

• Hardware implementation of lossless data compression is important for optimizing the capacity/cost/power of storage devices.
• GZIP is a file format used for file compression and decompression, which is based on one of the most popular algorithms for lossless storage - DEFLATE.
• DEFLATE uses a combination of the LZ77 algorithm and Huffman coding.

In this work we use the MaxCompiler programming tool suite to implement LZ77 algorithm on a FPGA.

Performance:

• Throughput: 8 input bytes per cycle clock
• Compression ratio: 1.88
• Compression speed: 1.6GB/s on FPGA with a clock frequency of 200MHz

Resources Usage:

• Logic resources: 255764/262400 (97.47%)
• Block memory: 1315/2567 (51.23%)

Folder Hierarchy(More info can be found in each folder)

• APP/
  • build/ (Where you can build/run DFE or simulations)
  • src/ (Where you can find all the source codes)
    • EngineCode/
    • CPUCode/
• TEST/
  • Calgary Corpus/ (The dataset that can be used)
• DOCS/ (The detailed documentations for the project)
• ORIG/ (An open-source software implementation of Lz77)
  • software/

• One can refer to the DOCS/GZip project_documentation for more information about the project(e.g. detailed explanations about the design, the limitations, and the options.)