[RADNEXT] Thermal neutron effects in electronic components

Thermal Neutrons Effects on Supercomputers and Autonomous Vehicles: Have we learned the lesson?

19 Sept 2022, 16:55

40m

Speaker

Paolo Rech (UFRGS)

Description

The high performance, high efficiency, and low cost of Commercial Off-The-Shelf (COTS) devices make them attractive for applications with strict reliability constraints. As a result, today COTS devices are adopted in HPC and safety-critical applications such as autonomous driving. Unfortunately, the cheap natural boron widely used in COTS chip manufacturing process makes them highly susceptible to thermal (low energy) neutrons. In this talk, we demonstrate, comparing the experimentally measured error rate to high and low energy neutrons, that thermal neutrons are still a significant threat to COTS device reliability. In the talk, to have a broad overview, we consider two DDR memories, an AMD APU, three NVIDIA GPUs, an Intel accelerator, and an FPGA executing a relevant set of algorithms. We predict the error rate of COTS in different scenarios that impact the thermal neutron flux such as weather, concrete walls and floors, and HPC liquid cooling systems. Correlating beam experiments and neutron detector data, we show that thermal neutrons FIT rate could be comparable or even higher than the high energy neutron FIT rate.