A Calibratable Model for Fast Energy Estimation of MVM Operations on RRAM Crossbars

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cubero-Cascante, José, Vaidyanathan, Arunkumar, Pelke, Rebecca, Pfeifer, Lorenzo, Leupers, Rainer, Joseph, Jan Moritz
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917663874220032
author Cubero-Cascante, José
Vaidyanathan, Arunkumar
Pelke, Rebecca
Pfeifer, Lorenzo
Leupers, Rainer
Joseph, Jan Moritz
author_facet Cubero-Cascante, José
Vaidyanathan, Arunkumar
Pelke, Rebecca
Pfeifer, Lorenzo
Leupers, Rainer
Joseph, Jan Moritz
contents The surge in AI usage demands innovative power reduction strategies. Novel Compute-in-Memory (CIM) architectures, leveraging advanced memory technologies, hold the potential for significantly lowering energy consumption by integrating storage with parallel Matrix-Vector-Multiplications (MVMs). This study addresses the 1T1R RRAM crossbar, a core component in numerous CIM architectures. We introduce an abstract model and a calibration methodology for estimating operational energy. Our tool condenses circuit-level behaviour into a few parameters, facilitating energy assessments for DNN workloads. Validation against low-level SPICE simulations demonstrates speedups of up to 1000x and energy estimations with errors below 1%.
format Preprint
id arxiv_https___arxiv_org_abs_2405_04326
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Calibratable Model for Fast Energy Estimation of MVM Operations on RRAM Crossbars
Cubero-Cascante, José
Vaidyanathan, Arunkumar
Pelke, Rebecca
Pfeifer, Lorenzo
Leupers, Rainer
Joseph, Jan Moritz
Signal Processing
C.3; I.2; I.6
The surge in AI usage demands innovative power reduction strategies. Novel Compute-in-Memory (CIM) architectures, leveraging advanced memory technologies, hold the potential for significantly lowering energy consumption by integrating storage with parallel Matrix-Vector-Multiplications (MVMs). This study addresses the 1T1R RRAM crossbar, a core component in numerous CIM architectures. We introduce an abstract model and a calibration methodology for estimating operational energy. Our tool condenses circuit-level behaviour into a few parameters, facilitating energy assessments for DNN workloads. Validation against low-level SPICE simulations demonstrates speedups of up to 1000x and energy estimations with errors below 1%.
title A Calibratable Model for Fast Energy Estimation of MVM Operations on RRAM Crossbars
topic Signal Processing
C.3; I.2; I.6
url https://arxiv.org/abs/2405.04326