Exploring the Potential of Wireless-enabled Multi-Chip AI Accelerators

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Irabor, Emmanuel, Musavi, Mariam, Das, Abhijit, Abadal, Sergi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918000715628544
author Irabor, Emmanuel
Musavi, Mariam
Das, Abhijit
Abadal, Sergi
author_facet Irabor, Emmanuel
Musavi, Mariam
Das, Abhijit
Abadal, Sergi
contents The insatiable appetite of Artificial Intelligence (AI) workloads for computing power is pushing the industry to develop faster and more efficient accelerators. The rigidity of custom hardware, however, conflicts with the need for scalable and versatile architectures capable of catering to the needs of the evolving and heterogeneous pool of Machine Learning (ML) models in the literature. In this context, multi-chiplet architectures assembling multiple (perhaps heterogeneous) accelerators are an appealing option that is unfortunately hindered by the still rigid and inefficient chip-to-chip interconnects. In this paper, we explore the potential of wireless technology as a complement to existing wired interconnects in this multi-chiplet approach. Using an evaluation framework from the state-of-the-art, we show that wireless interconnects can lead to speedups of 10% on average and 20% maximum. We also highlight the importance of load balancing between the wired and wireless interconnects, which will be further explored in future work.
format Preprint
id arxiv_https___arxiv_org_abs_2501_17567
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Exploring the Potential of Wireless-enabled Multi-Chip AI Accelerators
Irabor, Emmanuel
Musavi, Mariam
Das, Abhijit
Abadal, Sergi
Hardware Architecture
Artificial Intelligence
The insatiable appetite of Artificial Intelligence (AI) workloads for computing power is pushing the industry to develop faster and more efficient accelerators. The rigidity of custom hardware, however, conflicts with the need for scalable and versatile architectures capable of catering to the needs of the evolving and heterogeneous pool of Machine Learning (ML) models in the literature. In this context, multi-chiplet architectures assembling multiple (perhaps heterogeneous) accelerators are an appealing option that is unfortunately hindered by the still rigid and inefficient chip-to-chip interconnects. In this paper, we explore the potential of wireless technology as a complement to existing wired interconnects in this multi-chiplet approach. Using an evaluation framework from the state-of-the-art, we show that wireless interconnects can lead to speedups of 10% on average and 20% maximum. We also highlight the importance of load balancing between the wired and wireless interconnects, which will be further explored in future work.
title Exploring the Potential of Wireless-enabled Multi-Chip AI Accelerators
topic Hardware Architecture
Artificial Intelligence
url https://arxiv.org/abs/2501.17567