The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
MTJ를 적용한 비휘발성 메모리(NVM)는 읽기/쓰기 성능, 최고의 내구성, 표준 CMOS와의 작동 전압 호환성 등 많은 장점을 갖고 있습니다. 그러나 데이터를 쓸 때 많은 에너지를 소비합니다. 이는 배터리로 작동하는 모바일 기기에 적용할 때 걸림돌이 됩니다. 이 문제를 해결하기 위해 우리는 NVM에서 쓰기 작업에 대한 정밀 스케일링 기술의 성능을 향상시키는 접근 방식을 제안합니다. 정밀 스케일링은 에너지 절감을 위해 데이터의 비트 폭(예: 정밀도)을 줄이는 대략적인 컴퓨팅 기술입니다. 정밀 스케일링을 사용하여 이미지 데이터를 NVM에 쓸 때 쓰기 에너지와 이미지 품질은 쓰기 시간과 대상 비트 범위에 따라 변경됩니다. 우리는 비트 위치를 최적화하여 데이터를 분할하고 각 비트 범위에 대한 쓰기 시간을 최적화하여 데이터를 쓸 수 있는 비휘발성 플립플롭과 자기 랜덤 액세스 메모리(MRAM)를 위한 에너지 효율적인 근사 저장 방식을 제안합니다. . 통계 모델을 사용하여 쓰기 에너지 감소와 이미지 품질 저하 간의 균형을 유지하면서 쓰기 시간과 목표 비트 범위에 대한 최적의 값을 얻었습니다. 시뮬레이션 결과는 이러한 최적 값을 사용하면 허용 가능한 이미지 품질을 유지하면서 쓰기 에너지를 최대 50%까지 줄일 수 있음을 보여주었습니다. 또한 이 접근 방식을 자세히 사용할 때 입력 이미지와 출력 이미지 품질 간의 관계를 조사했습니다. 또한 선형 필터 및 에지 검출기를 포함한 12.5가지 유형의 이미지 처리에 우리의 접근 방식을 적용할 때 에너지 이점을 평가했습니다. 결과는 쓰기 에너지가 최대 XNUMX% 더 감소한 것으로 나타났습니다.
Yoshinori ONO
Shibaura Institute of Technology
Kimiyoshi USAMI
Shibaura Institute of Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
부
Yoshinori ONO, Kimiyoshi USAMI, "Energy Efficient Approximate Storing of Image Data for MTJ Based Non-Volatile Flip-Flops and MRAM" in IEICE TRANSACTIONS on Electronics,
vol. E104-C, no. 7, pp. 338-349, July 2021, doi: 10.1587/transele.2020CDP0009.
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 non-volatile flip-flops and a magnetic random-access memory (MRAM) 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 up to 50% while maintaining the acceptable image quality. We also investigated the relationship between the input images and the output image quality when using this approach in detail. In addition, we evaluated the energy benefits when applying our approach to nine types of image processing including linear filters and edge detectors. Results showed that the write energy is reduced by further 12.5% at the maximum.
URL: https://global.ieice.org/en_transactions/electronics/10.1587/transele.2020CDP0009/_p
부
@ARTICLE{e104-c_7_338,
author={Yoshinori ONO, Kimiyoshi USAMI, },
journal={IEICE TRANSACTIONS on Electronics},
title={Energy Efficient Approximate Storing of Image Data for MTJ Based Non-Volatile Flip-Flops and MRAM},
year={2021},
volume={E104-C},
number={7},
pages={338-349},
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 non-volatile flip-flops and a magnetic random-access memory (MRAM) 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 up to 50% while maintaining the acceptable image quality. We also investigated the relationship between the input images and the output image quality when using this approach in detail. In addition, we evaluated the energy benefits when applying our approach to nine types of image processing including linear filters and edge detectors. Results showed that the write energy is reduced by further 12.5% at the maximum.},
keywords={},
doi={10.1587/transele.2020CDP0009},
ISSN={1745-1353},
month={July},}
부
TY - JOUR
TI - Energy Efficient Approximate Storing of Image Data for MTJ Based Non-Volatile Flip-Flops and MRAM
T2 - IEICE TRANSACTIONS on Electronics
SP - 338
EP - 349
AU - Yoshinori ONO
AU - Kimiyoshi USAMI
PY - 2021
DO - 10.1587/transele.2020CDP0009
JO - IEICE TRANSACTIONS on Electronics
SN - 1745-1353
VL - E104-C
IS - 7
JA - IEICE TRANSACTIONS on Electronics
Y1 - July 2021
AB - 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 non-volatile flip-flops and a magnetic random-access memory (MRAM) 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 up to 50% while maintaining the acceptable image quality. We also investigated the relationship between the input images and the output image quality when using this approach in detail. In addition, we evaluated the energy benefits when applying our approach to nine types of image processing including linear filters and edge detectors. Results showed that the write energy is reduced by further 12.5% at the maximum.
ER -