Comparative analysis of soft-error sensitivity in LU decomposition algorithms on diverse GPUs

Abstract

Graphics processing units (GPUs) have become integral to embedded systems and supercomputing centres due to their large memory, cutting-edge technology and high performance per watt. However, their susceptibility to transient errors requires a comprehensive analysis of error sensitivity, as well as the development of error mitigation techniques and fault-tolerant algorithms. This study focuses on evaluating the soft-error sensitivity of two distinct versions of LU decomposition algorithms implemented on two very diferent GPUs—a low-power SoC embedded GPU and a high-performance massively parallel GPU. Through extensive fault injection campaigns on both GPUs, we examine the vulnerability of the algorithms, identify error causes, and determine critical code components requiring enhanced protection. The experiments reveal that most single bit fip fault injections in the instruction results lead to erroneous outcomes or unrecoverable errors. Notably, efcient GPU resource utilisation can increase the number of masked errors, thereby enhancing error resilience. Additionally, while diferent parts of the code exhibit similar error occurrence types and rates, the propagation of errors to elements within the result matrix difers signifcantly