Data Analytics and Hardware Acceleration - 2024.2 English

Vitis Libraries

Release Date
2024-11-29
Version
2024.2 English

Data analytics aims to accelerate a variety applications including data mining, text processing, formatting from raw data to dataframe, and querying operations on spatial data.

As an typical use case of data analysis, CSV Scanner is used to convert csv data into formatted data storage for subsequent analysis. An AMD Alveo™ card can help improve the CSV Scanner’s performance in the following way:

  1. Instruction parallelism by creating a customized and long pipeline.
  2. Data parallelism by processing multiple rows/files at the same time.
  3. Customizable memory hierarchy of BRAM/URAM/DDR, providing high bandwidth of memory access to help operators.