Saved in:
Bibliographic Details
Main Authors: Zhou, P. J., Yu, Q., Chen, M., Wang, Y. C., Meng, L. W., Zuo, Y., Ning, N., Liu, Y., Hu, S. G., Qiao, G. C.
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.01151
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929371132985344
author Zhou, P. J.
Yu, Q.
Chen, M.
Wang, Y. C.
Meng, L. W.
Zuo, Y.
Ning, N.
Liu, Y.
Hu, S. G.
Qiao, G. C.
author_facet Zhou, P. J.
Yu, Q.
Chen, M.
Wang, Y. C.
Meng, L. W.
Zuo, Y.
Ning, N.
Liu, Y.
Hu, S. G.
Qiao, G. C.
contents Edge-AI computing requires high energy efficiency, low power consumption, and relatively high flexibility and compact area, challenging the AI-chip design. This work presents a 0.96 pJ/SOP heterogeneous neuromorphic system-on-chip (SoC) with fullerene-like interconnection topology for edge-AI computing. The neuromorphic core integrates different technologies to augment computing energy efficiency, including sparse computing, partial membrane potential updates, and non-uniform weight quantization. Multiple neuromorphic cores and multi-mode routers form a fullerene-like network-on-chip (NoC). The average degree of communication nodes exceeds traditional topologies by 32%, with a minimal degree variance of 0.93, allowing advanced decentralized on-chip communication. Additionally, the NoC can be scaled up through extended off-chip high-level router nodes. A RISC-V CPU and a neuromorphic processor are tightly coupled and fabricated within a 5.42 mm^2 die area under 55 nm CMOS technology. The chip has a low power density of 0.52 mW/mm^2, reducing 67.5% compared to related works, and achieves a high neuron density of 30.23 K/mm^2. Eventually, the chip is demonstrated to be effective on different datasets and achieves 0.96 pJ/SOP energy efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2406_01151
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A 0.96pJ/SOP, 30.23K-neuron/mm^2 Heterogeneous Neuromorphic Chip With Fullerene-like Interconnection Topology for Edge-AI Computing
Zhou, P. J.
Yu, Q.
Chen, M.
Wang, Y. C.
Meng, L. W.
Zuo, Y.
Ning, N.
Liu, Y.
Hu, S. G.
Qiao, G. C.
Hardware Architecture
Edge-AI computing requires high energy efficiency, low power consumption, and relatively high flexibility and compact area, challenging the AI-chip design. This work presents a 0.96 pJ/SOP heterogeneous neuromorphic system-on-chip (SoC) with fullerene-like interconnection topology for edge-AI computing. The neuromorphic core integrates different technologies to augment computing energy efficiency, including sparse computing, partial membrane potential updates, and non-uniform weight quantization. Multiple neuromorphic cores and multi-mode routers form a fullerene-like network-on-chip (NoC). The average degree of communication nodes exceeds traditional topologies by 32%, with a minimal degree variance of 0.93, allowing advanced decentralized on-chip communication. Additionally, the NoC can be scaled up through extended off-chip high-level router nodes. A RISC-V CPU and a neuromorphic processor are tightly coupled and fabricated within a 5.42 mm^2 die area under 55 nm CMOS technology. The chip has a low power density of 0.52 mW/mm^2, reducing 67.5% compared to related works, and achieves a high neuron density of 30.23 K/mm^2. Eventually, the chip is demonstrated to be effective on different datasets and achieves 0.96 pJ/SOP energy efficiency.
title A 0.96pJ/SOP, 30.23K-neuron/mm^2 Heterogeneous Neuromorphic Chip With Fullerene-like Interconnection Topology for Edge-AI Computing
topic Hardware Architecture
url https://arxiv.org/abs/2406.01151