Evaluating the soft error sensitivity of a GPU-based SoC for matrix multiplication Articles uri icon

publication date

  • November 2020

start page

  • 1

end page

  • 5

volume

  • 114

International Standard Serial Number (ISSN)

  • 0026-2714

abstract

  • System-on-Chip (SoC) devices can be composed of low-power multicore processors combined with a smallgraphics accelerator (or GPU) which offers a trade-off between computational capacity and low-power consumption. In this work we use the LLFI-GPU fault injection tool on one of these devices to compare the sensitivity to soft errors of two different CUDA versions of matrix multiplication benchmark. Specifically, we perform faultinjection campaigns on a Jetson TK1 development kit, a board equipped with a SoC including an NVIDIA "Kepler" Graphics Processing Unit (GPU). We evaluate the effect of modifying the size of the problem and also the thread-block size on the behaviour of the algorithms. Our results show that the block version of the matrix multiplication benchmark that leverages the shared memory of the GPU is not only faster than the element-wiseversion, but it is also much more resilient to soft errors. We also use the cuda-gdb debugger to analyze the maincauses of the crashes in the code due to soft errors. Our experiments show that most of the errors are due toaccesses to invalid positions of the different memories of the GPU, which causes that the block version suffers ahigher percentage of this kind of errors.

keywords

  • fault injection; gpu; sensitivity; soft errors