Incident Library
API · MCP
한국어
Organisation
NERSC
NERSC ·
nersc
1 case · 2025
Root cause
Hardware failure
Component
GPU hardware and memory
Contributing factor
–
Symptom
GPU unavailable
Cases
IL-0482
Bursts of uncorrectable A100 memory errors on Perlmutter and Polaris reached over 1,000 double-bit errors a day
NERSC
· 2025
Root cause
Hardware failure
한국어