Graphtoy: Fast Software Simulation of Applications for AMD’s AI Engines
Embedded Systems and Applications, Technical University of Darmstadt, Darmstadt, Germany
Applied Reconfigurable Computing. Architectures, Tools, and Applications, 2024
@inproceedings{strobl2024arc,
author={Strobl, Jonathan and Solis-Vasquez, Leonardo and Lavan, Yannick and Koch, Andreas},
title={Graphtoy: Fast Software Simulation of Applications for AMD’s AI Engines},
booktitle={Applied Reconfigurable Computing. Architectures, Tools, and Applications},
year={2024},
publisher={Springer International Publishing}
}
This work presents Graphtoy, a coroutine-based compute graph simulator built in C++20, which can be embedded into a target application for rapid step-by-step prototyping of graphs targeting AMD’s AI Engines, as used in Versal FPGAs and Ryzen 7040 CPUs. By using a molecular docking application as a case study, we demonstrate: 1) how compute graphs developed using Graphtoy can be ported to the AI Engines with no modifications to the graph structure, and 2) that C++20 coroutines are well suited for simulating many-core systems with complex inter-core communication schemes. Furthermore, our set of molecular docking graphs ported to Graphtoy achieves an order-of-magnitude increase in simulation speed compared to AMD’s AI Engine graph simulators. The corresponding code is released as open source.
February 18, 2024 by hgpu
