Energy Efficient Approximate Storing of Image Data for MTJ Based Non-volatile Memory

Yoshinori Ono, Kimiyoshi Usami

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A non-volatile memory (NVM) employing MTJ has a lot of strong points such as read/write performance, best endurance and operating-voltage compatibility with standard CMOS. However, it consumes a lot of energy when writing the data. This becomes an obstacle when applying to battery-operated mobile devices. To solve this problem, we propose an approach to augment the capability of the precision scaling technique for the write operation in NVM. Precision scaling is an approximate computing technique to reduce the bit width of data (i.e. precision) for energy reduction. When writing image data to NVM with the precision scaling, the write energy and the image quality are changed according to the write time and the target bit range. We propose an energy-efficient approximate storing scheme for NVM that allows us to write the data by optimizing the bit positions to split the data and the write time for each bit range. By using the statistical model, we obtained optimal values for the write time and the targeted bit range under the trade-off between the write energy reduction and image quality degradation. Simulation results have demonstrated that by using these optimal values the write energy can be reduced by 25-50% while maintaining the acceptable image quality. In addition, we evaluated the energy benefits when applying our approach to three types of image processing. Results showed that the write energy is reduced by further 17% at the maximum.

Original languageEnglish
Title of host publicationProceedings - 9th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728184821
DOIs
Publication statusPublished - 2020 Aug
Event9th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2020 - Virtual, Seoul, Korea, Republic of
Duration: 2020 Aug 192020 Aug 21

Publication series

NameProceedings - 9th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2020

Conference

Conference9th IEEE Non-Volatile Memory Systems and Applications Symposium, NVMSA 2020
CountryKorea, Republic of
CityVirtual, Seoul
Period20/8/1920/8/21

Keywords

  • Approximate Computing
  • Image Data
  • Image Quality
  • Magnetic Tunnel Junction
  • Non-volatile Memory
  • Precision Scaling
  • Write Energy

ASJC Scopus subject areas

  • Hardware and Architecture

Fingerprint Dive into the research topics of 'Energy Efficient Approximate Storing of Image Data for MTJ Based Non-volatile Memory'. Together they form a unique fingerprint.

Cite this