AI Engine Components - 2025.2 English - UG1076

AI Engine Tools and Flows User Guide (UG1076)

Document ID
UG1076
Release Date
2025-11-20
Version
2025.2 English
AI Engine
Each AI Engine is a very long instruction word (VLIW) processor containing the following components.
  • A scalar unit
  • A vector unit
  • Two load units
  • A single store unit
AI Engine Tile
An AI Engine tile contains the following components.
  • An AI Engine
  • A local memory module together with several communication paths to support data exchange between tiles.
AI Engine Array
The AI Engine array consists of a 2D array of AI Engine tiles, where each AI Engine tile contains the following components.
  • An AI Engine
  • Memory module
  • A tile interconnect module
AI Engine Program
Consists of a data flow graph specification, written as a C++ class. A top-level program instantiates one or more graph classes and is compiled and executed using the AI Engine tool chain.
AI Engine Kernels
AI Engine vector data types and intrinsic functions are used to write kernels in C/C++. These are the computation functions running on an AI Engine. The kernels form the fundamental building blocks of a data flow graph specification.
ADF Graph
An ADF graph is a network with a single AI Engine kernel or multiple AI Engine kernels connected by data streams. The ADF graph interacts with the programmable logic, global memory, and processing system with specific constructs such as RTP and the following.
PLIO
The port attribute in the graph that makes stream connections to or from the programmable logic.
GMIO
The port attribute in the graph that is makes external memory-mapped connections to or from the global memory.
RTP
The Runtime Parameters (RTP) are communication channels between the processing system (PS) and the AI Engine. RTPs are suitable for low-throughput and low-latency signals.

You can embed graphs into other graphs as sub-graphs, allowing you to structure the design into logical subsystems.