OISMA: On-the-fly In-memory Stochastic Multiplication Architecture for Matrix-Multiplication Workloads

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Agwa, Shady, Pan, Yihan, Papandroulidakis, Georgios, Prodromakis, Themis
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913054094000128
author Agwa, Shady
Pan, Yihan
Papandroulidakis, Georgios
Prodromakis, Themis
author_facet Agwa, Shady
Pan, Yihan
Papandroulidakis, Georgios
Prodromakis, Themis
contents Artificial intelligence (AI) models are currently driven by a significant upscaling of their complexity, with massive matrix-multiplication workloads representing the major computational bottleneck. In-memory computing (IMC) architectures are proposed to avoid the von Neumann bottleneck. However, both digital/binary-based and analog IMC architectures suffer from various limitations, which significantly degrade the performance and energy efficiency gains. This work proposes OISMA, an energy-efficient IMC architecture that utilizes the computational simplicity of a quasi-stochastic computing (SC) domain (bent-pyramid (BP) system) while keeping the same efficiency, scalability, and productivity of digital memories. OISMA converts normal memory read operations into in situ stochastic multiplication operations with a negligible cost. An accumulation periphery then accumulates the output multiplication bitstreams, achieving the matrix multiplication (MatMul) functionality. A 4-kB 1T1R OISMA array was implemented using a commercial 180-nm technology node and in-house resistive random-access memory (RRAM) technology. At 50 MHz, it achieves 0.789 TOPS/W and 3.98 GOPS/mm2 for energy and area efficiency, respectively, occupying an effective computing area of 0.804241 mm2. Scaling OISMA to 22-nm technology shows a significant improvement of two orders of magnitude in energy efficiency and one order of magnitude in area efficiency, compared to dense MatMul IMC architectures.
format Preprint
id arxiv_https___arxiv_org_abs_2508_08822
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle OISMA: On-the-fly In-memory Stochastic Multiplication Architecture for Matrix-Multiplication Workloads
Agwa, Shady
Pan, Yihan
Papandroulidakis, Georgios
Prodromakis, Themis
Hardware Architecture
Artificial Intelligence
Emerging Technologies
Performance
Artificial intelligence (AI) models are currently driven by a significant upscaling of their complexity, with massive matrix-multiplication workloads representing the major computational bottleneck. In-memory computing (IMC) architectures are proposed to avoid the von Neumann bottleneck. However, both digital/binary-based and analog IMC architectures suffer from various limitations, which significantly degrade the performance and energy efficiency gains. This work proposes OISMA, an energy-efficient IMC architecture that utilizes the computational simplicity of a quasi-stochastic computing (SC) domain (bent-pyramid (BP) system) while keeping the same efficiency, scalability, and productivity of digital memories. OISMA converts normal memory read operations into in situ stochastic multiplication operations with a negligible cost. An accumulation periphery then accumulates the output multiplication bitstreams, achieving the matrix multiplication (MatMul) functionality. A 4-kB 1T1R OISMA array was implemented using a commercial 180-nm technology node and in-house resistive random-access memory (RRAM) technology. At 50 MHz, it achieves 0.789 TOPS/W and 3.98 GOPS/mm2 for energy and area efficiency, respectively, occupying an effective computing area of 0.804241 mm2. Scaling OISMA to 22-nm technology shows a significant improvement of two orders of magnitude in energy efficiency and one order of magnitude in area efficiency, compared to dense MatMul IMC architectures.
title OISMA: On-the-fly In-memory Stochastic Multiplication Architecture for Matrix-Multiplication Workloads
topic Hardware Architecture
Artificial Intelligence
Emerging Technologies
Performance
url https://arxiv.org/abs/2508.08822