A Latency-Constrained, Gated Recurrent Unit (GRU) Implementation in the Versal AI Engine

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sapkas, M., Triossi, A., Zanetti, M.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917240280973312
author Sapkas, M.
Triossi, A.
Zanetti, M.
author_facet Sapkas, M.
Triossi, A.
Zanetti, M.
contents This work explores the use of the AMD Xilinx Versal Adaptable Intelligent Engine (AIE) to accelerate Gated Recurrent Unit (GRU) inference for latency constrained applications. We present a custom workload distribution framework across the AIE's vector processors and propose a hybrid AIE - Programmable Logic (PL) design to optimize computational efficiency. Our approach explores the parallelization over the rows of the matrices by utilizing as many of the AIE vectorized processors effectively computing all the elements of the resulting vector at the same time, an alternative to cascade stream pipelining.
format Preprint
id arxiv_https___arxiv_org_abs_2511_15626
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Latency-Constrained, Gated Recurrent Unit (GRU) Implementation in the Versal AI Engine
Sapkas, M.
Triossi, A.
Zanetti, M.
Performance
This work explores the use of the AMD Xilinx Versal Adaptable Intelligent Engine (AIE) to accelerate Gated Recurrent Unit (GRU) inference for latency constrained applications. We present a custom workload distribution framework across the AIE's vector processors and propose a hybrid AIE - Programmable Logic (PL) design to optimize computational efficiency. Our approach explores the parallelization over the rows of the matrices by utilizing as many of the AIE vectorized processors effectively computing all the elements of the resulting vector at the same time, an alternative to cascade stream pipelining.
title A Latency-Constrained, Gated Recurrent Unit (GRU) Implementation in the Versal AI Engine
topic Performance
url https://arxiv.org/abs/2511.15626