Single Event Effect Evaluation of the Jetson AGX Xavier Module Using Proton Irradiation

Commercial Off-The-Shelf (COTS) System-on-Chip (SoC) are becoming widespread in embedded systems. Many of them include a multicore CPU and a high-end GPU. They combine high computational performance with low power consumption and flexible multilevel parallelism. This kind of device is also being considered for radiation environments where large amounts of data must be processed or compute intensive applications must be executed. In this paper we compare three different strategies to perform matrix multiplication in the GPU of a Tegra TK1 SoC. Our aim is to analyze how the different use of the resources of the GPU influences, not only the computational performance of the algorithm, but also its radiation sensitivity. Radiation experiments with protons were performed to compare the behaviour of the three strategies. Experimental results show that most of the errors force a reboot of the platform. The number of errors is directly related with how the algorithms use the internal memories of the GPU, and increases with the matrix size. It is also related with the number of transactions with the global memory, which in our experiments is not affected by the radiation. Results show that the smallest cross-section is obtained with the fastest algorithm, even if it uses the cores of the GPU more intensively.

Section: Related Worksupporting

confidence: 78%

Comparison of Parallel Implementation Strategies in GPU-Accelerated System-on-Chip Under Proton Irradiation

Badía

León

Belloch

et al. 2022

“…Preliminary results suggest operation beyond 20 krad(Si). Similar conclusions were reached in [26] for a Jetson AGX Xavier board, including a 512-core NVIDIA GPU, using proton irradiation. High-energy protons were used in [10] to evaluate the cross-section and the total dose performance of the Tegra K1 SoC by running different versions of the FFT in the CPU and GPU.…”

Section: Related Worksupporting

confidence: 78%

Reliability Evaluation of LU Decomposition on GPU-Accelerated System-on-Chip Under Proton Irradiation

Badía

León

Belloch

et al. 2022

Graphic Processing Units (GPU) have become a basic accelerator both in high-performance nodes and low-power SoC. They provide massive data parallelism and very high performance per watt. However, their reliability in harsh environments is an important issue to take into account, especially for safetycritical applications. In this paper we evaluate the influence of the parallelization strategy on reliability of LU decomposition on a GPU-accelerated SoC under proton irradiation. Specifically we compare a memory bound and a compute bound implementation of the decomposition on a K20A GPU embedded on a TK1 SoC. We leverage the GPU and CPU clock frequencies both to highlight the radiation sensitivity of the GPU where we are running the benchmark, and also to apply both algorithms to solve problems with the same size when exposed to the same radiation dose. Results show that a more intensive use of the resources of the GPU increases the cross-section. We also observed that most of the radiation-induced errors hang the operating system and even the rebooting process. Finally, we present a preliminary study of the error propagation of the LU decomposition algorithms.

“…The mitigation technique under study is partitioned between the GPU and the CPU and its effect on the overall reliability of the SoC system was investigated. Previous research has mainly used desktop GPUs to analyze their sensitivity to radiation in isolation [9], [10] while recent studies on GPU-accelerated SoCs independently examined the GPU and CPU components (e.g., using TMR solely on the CPUs to eliminate their contribution to radiationinduced faults) [11]. Our research helps to establish the importance of each part to improve the mitigation technique conveniently.…”

Section: Introductionmentioning

confidence: 90%

Analysis of Kernel Redundancy for Soft Error Mitigation on Embedded GPUs

Serrano-Cases

Alcaide

Romero-Maestre

et al. 2023

The use of state-of-the-art commercial processors such as graphical processing units (GPUs) is becoming increasingly common in the New Space industry in order to ensure high performance and power efficiency. However, commercial GPUs are not designed to operate in a harsh environment and therefore different protection techniques need to be applied to mitigate the effects of radiation, including those produced by single events. This paper assesses the effectiveness of redundant kernel execution on tightly constrained embedded GPUs under proton irradiation, with results suggesting a significant improvement in the SDC cross-section without penalizing the stability of the whole system. In addition, the posterior error analysis shows that the CPU is the source of the majority of the events, which are mainly dominated by functional interrupts