Overview
compression_tile_graphic

With the explosion of data that needs to be both transferred and processed between different end-points at the edge and in your data centers, the ability to efficiently compress the data affects the real-time performance and storage requirements. This affects the bottom line profitability of your solutions.

CPU-powered data compression is unable to deliver the real-time performance demanded by today’s applications while keeping the storage and infrastructure costs low.

Vitis™ Data Compression library is a performance-optimized library to accelerate the Lempel-Ziv (LZ) data compression and decompression algorithms on Xilinx Accelerator cards. It is designed as a specialized compression engine, multiple of which can run concurrently on the same Xilinx accelerator card to meet the high-throughput requirements of your algorithms. This reduces the bandwidth consumption and the overall infrastructure costs, on-premise or in the cloud.

You can use the pre-optimized library kernels for LZ4 and Snappy compression/decompression or use the low-level optimized primitives as components while designing your end-to-end accelerated kernel.


Performance Benchmark

vitis-data-compression
Notes:
  • Xilinx Alveo U200 at 44% Utilization
  • CPU Corei7 9700K CPU @4.9GHz (single threaded) : Source
Getting Started