Algorithm Characteristics - 2024.1 English

Versal Adaptive SoC System and Solution Planning Methodology Guide (UG1504)

Document ID
UG1504
Release Date
2024-06-19
Version
2024.1 English

Algorithms can typically be characterized in several different dimensions such as:

  • Sampling Rates: A primary driver of design complexity and cost, both in terms of resource requirements for compute, storage, and bandwidth.
  • Data Types: A primary driver of design complexity and cost.
  • Storage Requirements: Buffering of data, storage of intermediate results, look-up tables, etc.
  • Compute Requirements: Multiply-accumulate per cycle.
  • Nature of Processing: Style of data flow graph, regular with no branching or data dependent selection vs. data dependent, block vs. sample based, word vs. bit oriented, pipelined vs. feedback.
  • Regularized DSP structures like filters and FFTs vs. specialized functions (Bessel function, Taylor series vs. LUT implementation).