SRAM On-Chip Monitoring Methodology for High Yield and Energy Efficient Memory Operation at Near Threshold Voltage
IEEE Computer Society Annual Symposium on VLSI (ISVLSI)(2019)
Seoul Natl Univ | Samsung Elect Co Ltd
Abstract
Low power design by near-threshold voltage (NTV) operation is very attractive since it affords to considerably mitigate the sharp increase of power dissipation. However, one key barrier for the use of NTV operation is the significant increase of the SRAM failure. In this work, we propose an on-chip SRAM monitoring methodology that is able to accurately predict the minimum voltage, Vddmin, on each die that does not cause SRAM failure under a target confidence level. Precisely, we propose an SRAM monitor, from which we measure a maximum voltage, Vfail that causes functional failure on that SRAM monitor. Then, we propose a novel methodology of inferring SRAM Vddmin on each die from the measured Vfail of SRAM monitor on the same die where Vfail-Vddmin correlation table is built-up in design infra development phase, and Vddmin can be directly derived from the measured Vfail referencing the correlation table in silicon production phase. Through experiments, we confirm that our proposed methodology is able to save leakage power by 7.43%, read energy by 3.98%, and write energy by 4.06% in SRAM bitcell array over that by applying a uniform minimum voltage for all dies while meeting the same yield constraint.
MoreTranslated text
Key words
SRAM failure,On chip monitoring methodology,Near threshold voltage,Low power,Process variation
求助PDF
上传PDF
View via Publisher
AI Read Science
AI Summary
AI Summary is the key point extracted automatically understanding the full text of the paper, including the background, methods, results, conclusions, icons and other key content, so that you can get the outline of the paper at a glance.
Example
Background
Key content
Introduction
Methods
Results
Related work
Fund
Key content
- Pretraining has recently greatly promoted the development of natural language processing (NLP)
- We show that M6 outperforms the baselines in multimodal downstream tasks, and the large M6 with 10 parameters can reach a better performance
- We propose a method called M6 that is able to process information of multiple modalities and perform both single-modal and cross-modal understanding and generation
- The model is scaled to large model with 10 billion parameters with sophisticated deployment, and the 10 -parameter M6-large is the largest pretrained model in Chinese
- Experimental results show that our proposed M6 outperforms the baseline in a number of downstream tasks concerning both single modality and multiple modalities We will continue the pretraining of extremely large models by increasing data to explore the limit of its performance
Upload PDF to Generate Summary
Must-Reading Tree
Example

Generate MRT to find the research sequence of this paper
Related Papers
Ultra Low Power Offering 14 Nm Bulk Double Gate FinFET Based SRAM Cells
Sustainable Computing Informatics and Systems 2022
被引用3
Data Disclaimer
The page data are from open Internet sources, cooperative publishers and automatic analysis results through AI technology. We do not make any commitments and guarantees for the validity, accuracy, correctness, reliability, completeness and timeliness of the page data. If you have any questions, please contact us by email: report@aminer.cn
Chat Paper