2010/01/01 by Imran S. Haque, Vijay S. Pande · 1 citation
Engineering · Computer Science · #Radiation Effects in Electronics #Parallel Computing and Optimization Techniques #Distributed systems and fault tolerance #Computer science #General-purpose computing on graphics processing units #Graphics #Soft error #Word error rate #Error detection and correction #Graphics processing unit #Reliability (semiconductor) #Parallel computing #Folding (DSP implementation) #CUDA #Algorithm #Computer graphics (images) #Artificial intelligence
paper · doi:10.1109/ccgrid.2010.84
openalex publication_date 2010/01/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/29
Graphics processing units (GPUs) are gaining widespread use in high-performance computing because of their performance advantages relative to CPUs. However, the reliability of GPUs is largely unproven. In particular, current GPUs lack error checking and correcting (ECC) in their memory subsystems. The impact of this design has not been previously measured at a large enough scale to quantify soft error events. We present MemtestG80, our software for assessing memory error rates on NVIDIA graphics cards. Furthermore, we present a large-scale assessment of GPU error rate, conducted by running MemtestG80 on over 50,000 hosts on the Folding@home distributed computing network. Our control experiments on consumer-grade and dedicated-GPGPU hardware in a controlled environment found no errors. However, our survey on Folding@home finds that, in their installed environments, two-thirds of tested GPUs exhibit a detectable, pattern-sensitive rate of memory soft errors. We show that these errors persist after controlling for over clocking and environmental proxies for temperature, but depend strongly on board architecture.