Skip to main navigation Skip to search Skip to main content

Software-Hardware Co-Design for Feature Extraction on Racetrack Memory-Based PIM

  • East China Normal University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

CNNs, which involve intensive matrix computations, often face memory bandwidth bottlenecks that PIM architectures aim to overcome. RM, with its high density, low power consumption, and excellent endurance, is particularly suited for storing matrix data in such architectures. We propose a software-level data placement strategy, tile blocking, designed to optimize the shift-based access mechanism inherent in RM. We integrate this strategy into a hardware-software co-design framework for CNN feature extraction. Our approach enhances data parallelism and significantly reduces the number of shift operations, thereby achieving energy-efficient computation on RM-based PIM systems. Experimental results demonstrate that our strategy reduced the average number of shifts by approximately 81.64%, reduced the average energy by approximately 38.68%, and reduced the average execution time by approximately 44.7%.

Original languageEnglish
Title of host publicationProceedings - 2025 14th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages57-62
Number of pages6
ISBN (Electronic)9798331585273
DOIs
StatePublished - 2025
Event2025 14th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2025 - Singapore, Singapore
Duration: 20 Aug 202522 Aug 2025

Publication series

NameProceedings - 2025 14th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2025

Conference

Conference2025 14th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2025
Country/TerritorySingapore
CitySingapore
Period20/08/2522/08/25

Keywords

  • Process-In-Memory
  • Racetrack Memory
  • Software-Hardware Co-Design

Fingerprint

Dive into the research topics of 'Software-Hardware Co-Design for Feature Extraction on Racetrack Memory-Based PIM'. Together they form a unique fingerprint.

Cite this