Single event upset failure probability evaluation and periodic scrubbing techniques for hierarchical parallel vision processors

LETTER

Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences">Qian Di, Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences">Zhongxing Zhang, Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences">Honglong Li, Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences">Zhao Zhang, Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences">Peng Feng, Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences
Center for Excellence in Brain Science and Intelligence Technology, Chinese Academy of Sciences">Nanjian Wu

Author information

Qian Di
State Key Laboratory of Superlattices and Microstructures, Institute of Semiconductors, Chinese Academy of Sciences
Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences
Zhongxing Zhang
State Key Laboratory of Superlattices and Microstructures, Institute of Semiconductors, Chinese Academy of Sciences
Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences
Honglong Li
State Key Laboratory of Superlattices and Microstructures, Institute of Semiconductors, Chinese Academy of Sciences
Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences
Zhao Zhang
State Key Laboratory of Superlattices and Microstructures, Institute of Semiconductors, Chinese Academy of Sciences
Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences
Peng Feng
State Key Laboratory of Superlattices and Microstructures, Institute of Semiconductors, Chinese Academy of Sciences
Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences
Nanjian Wu
State Key Laboratory of Superlattices and Microstructures, Institute of Semiconductors, Chinese Academy of Sciences
Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences
Center for Excellence in Brain Science and Intelligence Technology, Chinese Academy of Sciences

Keywords: single event upset, vision processor, fault injection, periodic scrubbing, availability

JOURNAL FREE ACCESS

2019 Volume 16 Issue 21 Pages 20190544

DOI https://doi.org/10.1587/elex.16.20190544

Browse “Advance online publication” version

Details

Abstract

This paper proposes novel single event upset (SEU) failure probability evaluation and periodic scrubbing techniques for hierarchical parallel vision processors. To automatically evaluate the SEU failure probability and identify all the critical elements in a processor, complementary fault injection methods based on logic circuit simulator and Perl script are proposed. These methods can be used to randomly inject faults into D flip-flops (DFFs) and various types of memory at the register transfer level (RTL) as well as to evaluate the vision processor performance. Based on the evaluation results, an accurate periodic scrubbing technique is proposed to increase the processor availability. The results denote that the peak availability of the processor over a period of one year can be improved from 18% to 99.9% after scrubbing the RISC program memory for a period of 10⁴ s. Therefore, we can improve the fault-tolerance performance of a vision processor while avoiding unnecessary area and power costs using techniques ranging from evaluation to mitigation.

Corresponding author

Register with J-STAGE for free!