Systematic Reliability Evaluation of FPGA Implemented CNN Accelerators

Abstract

Convolutional neural networks (CNN) have become essential for many scientific and industrial applications, such as image classification and pattern detection. Among the devices that can implement neural networks, SRAM based FPGAs are a popular option due to their excellent parallel computing capability and good flexibility. However, SRAM-FPGAs are susceptible to radiation effects, which limits its application on safety critical applications. In this paper, the reliability of an accelerator based on the advanced Instruction-Set Architecture is evaluated based on hardware fault injection experiments. Each main module of the accelerator is evaluated separately, and the impact of parallelism and model features on the accelerator reliability is also examined. The experimental results reveal some important conclusions in terms of general hardware reliability and also of the particular model reliability. First, over 99% of SEUs on the computation modules will cause accuracy loss, and the reliability improves for higher parallelism. Second, a large portion of SEUs on the data mover and the instruction scheduler will cause system corruptions due to abnormal interactions with the ARM or other modules. Third, nonlinear activation and pooling layers are effective in reducing the effect of SEUs on computation modules, so models that use these layers tend to be more robust. The results provide a deep understanding of the impact of errors on CNNs implemented on ISA based FPGA accelerators (e.g., the Xilinx DPU).

Artificial Neural Networks for Space and Safety-Critical Applications: Reliability Issues and Potential Solutions

“…When SRAM-based FPGA are considered, fault injection becomes a primary evaluation tool. In fact, since the configuration memory is accessible and represents the most critical resources of the device, in FPGAs fault injection can draw accurate evaluations [153], [154] and is highly suggested for machine learning applications.…”

Single event effects evaluation on convolution neural network in Xilinx 28 nm system on chip

“…[15] Moreover, attention has been given to the radiation sensitivity of different CNN models, such as failure distributions and mechanisms. [16] Furthermore, the impact of data bit compression or model quantization techniques on the SEE reliability of the CNN system has been evaluated. [17] In this paper, the CNN-based recognition of handwritten digits in MNIST data set has been implemented in Xilinx Zynq-7020 SoC, and its soft errors have been evaluated by alpha radiation sources and fault injection in CRAM.…”