Gzip decompression method, device and system based on FPGA

Abstract

The invention discloses a Gzip decompression method, device and system based on an FPGA and a server, and the method comprises the steps: a processor analyzing a data block head in a compressed data block in an obtained Gzip compressed file, and determining a coding mode of the compressed data block; if the encoding mode is a dynamic Huffman encoding mode, sending each to-be-decompressed data block and the respective corresponding dynamic Huffman code table and data block information to the FPGA so as to perform parallel decoding on the to-be-decompressed data blocks by utilizing the FPGA. According to the invention, Gzip decompression is realized through combination of the CPU and the FPGA, and parallel decoding is carried out on the to-be-decompressed data block by utilizing the FPGA according to the Huffman code table obtained through analysis of the processor, so that the Gzip decompression performance is improved, and the utilization rate of the processor is reduced. The processorgenerates the dynamic Huffman code table by adopting a normal form Huffman coding mode, so that the required code table quantity is reduced, and the decoding speed is further improved.

Description

technical field [0001] The invention relates to the field of data compression, in particular to an FPGA-based Gzip decompression method, device, system and server. Background technique [0002] With the advancement of information technology and the rapid development of the Internet industry, people are generating a large amount of data information every day, and the era of big data has come. The large-scale data volume has brought great challenges to data storage and network transmission bandwidth, so it is necessary to perform data compression under the condition of limited hardware resources. Data compression can be divided into lossy compression and lossless compression. Lossy compression will cause a certain degree of data distortion, and the original data cannot be reconstructed after decompression. Lossless compression is lossless compression, which only removes the redundancy of data during the compression process, and the data can be completely restored after decom...

AI Technical Summary

Problems solved by technology

However, most of the current servers use software to implement Gzip compression and decompression algorithms. There are two defects in this implementation: the one is that the CPU (processor) utilization rate will be very high, which will affect the operation of other programs on the server; the other is decompression. The algorithm requires repeated table lookup and memory read and write operations, which can only be implemented serially with the CPU, resulting in low execution efficiency and low decompression performance

Images