Optimized Distance Calculation Support for HBM PIM(Processing-In-Memory)

  • Nahyeon Kim
  • , Sujin Kim
  • , Haechannuri Noh
  • , Min Jung
  • , Huijin Roh
  • , Ji Hoon Kim

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

Abstract

In this paper, we propose an extension of the Processing-In-Memory (PIM) architecture's instruction set to optimize distance calculations. The existing PIM architecture, while efficient for neural network tasks, exhibits limitations in handling distance calculations due to frequent data transfers between memory banks. To address this, we introduced the custom instruction, reducing intermediate data storage needs and improving computation efficiency. Simulations using random matrices showed that our approach decreases cycle count by up to 44 %, with more significant performance gains for larger dataset. This extension demonstrates the potential of enhanced PIM architectures for efficient memory-bound operations.

Original languageEnglish
Title of host publication2025 International Conference on Electronics, Information, and Communication, ICEIC 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798331510756
DOIs
StatePublished - 2025
Event2025 International Conference on Electronics, Information, and Communication, ICEIC 2025 - Osaka, Japan
Duration: 19 Jan 202522 Jan 2025

Publication series

Name2025 International Conference on Electronics, Information, and Communication, ICEIC 2025

Conference

Conference2025 International Conference on Electronics, Information, and Communication, ICEIC 2025
Country/TerritoryJapan
CityOsaka
Period19/01/2522/01/25

Bibliographical note

Publisher Copyright:
© 2025 IEEE.

Keywords

  • PIM Simulator
  • Processing In Memory
  • vector search

Fingerprint

Dive into the research topics of 'Optimized Distance Calculation Support for HBM PIM(Processing-In-Memory)'. Together they form a unique fingerprint.

Cite this