High throughput FPGA implementation of LZ4 algorithm
Implementace kompresního algoritmu LZ4 s vysokou propustností v FPGA
Authors
Supervisors
Reviewers
Editors
Other contributors
Journal Title
Journal ISSN
Volume Title
Publisher
České vysoké učení technické v Praze
Czech Technical University in Prague
Czech Technical University in Prague
Date
Abstract
Diplomová práce se zabývá návrhem a implementací kompresní a dekompresní architektury číslicových jednotek určených pro FPGA obvody. Návrh klade důraz na využití v systémech s vysokou propustností a nízkou latencí. Práce obsahuje důkladnou analýzu kompresních algoritmů rodiny LZ77 a LZ78 pro dosáhnutí optimalizované implementace algoritmu LZ4 pro hardwarovou architekturu. Dále práce popisuje návrh kompresní jednotky v jazyce VHDL. Poslední část práce se věnuje simulaci, testování a experimentálnímu vyhodnocení navrhnuté jednotky. Navhrnutá architektura byla úspěšně implementována, simulována a otestována pomocí Ethernetového rozhraní na FPGA platformě od firmy Xilinx
This master thesis presents a design and an implementation of hardware compression and decompression units designated for use in FPGAs. The design focuses on high-throughput and low latency systems. The thesis contains a thorough analysis of LZ77 and LZ78 families of compression algorithms for implementation of optimized LZ4 algorithm for hardware architecture. Then it describes the design process of the compression unit written in VHDL. Lastly, it concerns with simulation, testing, and experimental evaluation of the designed architecture. The architecture has been successfully implemented, simulated and tested using Ethernet interface on the Xilinx FPGA platform.
This master thesis presents a design and an implementation of hardware compression and decompression units designated for use in FPGAs. The design focuses on high-throughput and low latency systems. The thesis contains a thorough analysis of LZ77 and LZ78 families of compression algorithms for implementation of optimized LZ4 algorithm for hardware architecture. Then it describes the design process of the compression unit written in VHDL. Lastly, it concerns with simulation, testing, and experimental evaluation of the designed architecture. The architecture has been successfully implemented, simulated and tested using Ethernet interface on the Xilinx FPGA platform.
Description
Keywords
Komprese, Decomprese, LZ4, Nízká latence, 1Gbit, 10Gbit, IP Core, FPGA, Compression, Decompression, LZ4, Low latency, 1Gbit, 10Gbit, IP Core, FPGA