Soft-Error Characterization and Mitigation Strategies for Edge Tensor Processing Units in Space

The Google Coral Edge Tensor Processing Unit (Edge TPU) offers low-power, high-performance capabilities ideal for enabling deep learning in space. However, as a commercial product, no reliability considerations are made in its design. As a device targeted by current and future space computing platforms, it is vital to mission success to understand the vulnerabilities and possible failure modes prior to flight. In this research, we evaluate the soft-error vulnerabilities of the Edge TPU and propose fault-mitigation techniques to improve device reliability. Several Edge TPUs were irradiated using a wide spectrum neutron beam at the Los Alamos Neutron Science Center to evaluate the reliability of two machine-learning applications with common use cases within the space domain: image classification and semantic segmentation. Through experimentation, a vulnerability within the onboard memory is identified. Responsible for caching model parameters for increased performance, the onboard memory represents a critical device area. Any upsets within the cache risk compromising data integrity and model determinism. Across a variety of models tested, fault accumulation and persistence are consistently observed, resulting in the degradation of model accuracy and confidence. To alleviate the impact of radiation, we propose two fault-mitigation techniques: Naive Refreshing (NR) and Golden Batch Refreshing (GBR). NR periodically reloads model parameters to clear corrupted data. GBR is proposed as an alternative method to reduce reload frequency and improve performance. By leveraging knowledge of the cache vulnerabilities and applying one or more mitigation strategies, Edge TPUs can be properly considered for integration into existing and future flight hardware.