Evaluation of Xilinx Deep Learning Processing Unit under Neutron Irradiation

2022/06/04 by Agiakatsikas, D., Foutris, N., Sari, A. +6
#Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #FOS: Physical sciences #Hardware Architecture (cs.AR) #Instrumentation and Detectors (physics.ins-det)

paper · doi:10.48550/arxiv.2206.01981

Abstract

This paper studies the dependability of the Xilinx Deep-Learning Processing Unit (DPU) under neutron irradiation. It analyses the impact of Single Event Effects (SEEs) on the accuracy of the DPU running the resnet50 model on a Xilinx Ultrascale+ MPSoC.

Related

• Single Event Effects Assessment of UltraScale+ MPSoC Systems under Atmospheric Radiation
• Experimental evaluation of neutron-induced errors on a multicore RISC-V platform
• Characterizing a Neutron-Induced Fault Model for Deep Neural Networks
• Analysis of Single Event Induced Bit Faults in a Deep Neural Network Accelerator Pipeline
• Mitigating multiple single-event upsets during deep neural network inference using fault-aware training