2021 IEEE International Parallel and Distributed Processing Symposium (IPDPS)
Demystifying GPU Reliability: Comparing and Combining Beam Experiments, Fault Simulation, and Profiling
DOI Bookmark: 10.1109/IPDPS49936.2021.00037
Authors
Fernando Fernandes dos Santos, UFRGS,BrazilSiva Kumar Sastry Hari, NVIDIA Corporation,United States
Pedro Martins Basso, UFRGS,Brazil
Luigi Carro, UFRGS,Brazil
Paolo Rech, Politecnico di Torino,Italy