Story of Two GPUs: Characterizing the Resilience of Hopper H100 and Ampere A100 GPUs

This study characterizes GPU resilience in Delta, a large-scale AI system that consists of 1,056 A100 and H100 GPUs, with over 1,300 petaflops of peak throughput. We used 2.5 years of operational data (11.7 million GPU hours) on GPU errors. Our major findings include: (i) H100 GPU memory resilience is worse than A100 GPU memory, with 3.2x lower per-GPU MTBE for memory errors, (ii) The GPU memory error-recovery mechanisms on H100 GPUs are insufficient to handle the increased memory capacity, (iii) H100 GPUs demonstrate significantly improved GPU hardware resilience over A100 GPUs with respect to critical hardware components, (iv) GPU errors on both A100 and H100 GPUs frequently result in job failures due to the lack of robust recovery mechanisms at the application level, and (v) We project the impact of GPU node availability on larger-scales and find that significant overprovisioning of 5% is necessary to handle GPU failures.

Authors

Institutions

Publication Details

Published
2025-11-12
DOI
https://doi.org/10.1145/3712285.3759821
Citations
1
Primary Topic
Radiation Effects in Electronics
Type
article
Field-Weighted Citation Impact
0.73
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Story of Two GPUs: Characterizing the Resilience of Hopper H100 and Ampere A100 GPUs

Phuong Cao, Saurabh Jha, Archit Patke, Thai-Hung Nguyen et al.
1 citations
Radiation Effects in Electronics
0.73
article

Story of Two GPUs: Characterizing the Resilience of Hopper H100 and Ampere A100 GPUs

Phuong Cao, Saurabh Jha, Archit Patke, Thai-Hung Nguyen, Daby Sow, Chandra Narayanaswami, Zbigniew Kalbarczyk, Shengkun Cui, Brett Bode, Ravishankar K. Iyer, Gregory H. Bauer, Aditya Ranjan, Ziheng Chen, Catello Di Martino
article en
1 citations

Abstract

This study characterizes GPU resilience in Delta, a large-scale AI system that consists of 1,056 A100 and H100 GPUs, with over 1,300 petaflops of peak throughput. We used 2.5 years of operational data (11.7 million GPU hours) on GPU errors. Our major findings include: (i) H100 GPU memory resilience is worse than A100 GPU memory, with 3.2x lower per-GPU MTBE for memory errors, (ii) The GPU memory error-recovery mechanisms on H100 GPUs are insufficient to handle the increased memory capacity, (iii) H100 GPUs demonstrate significantly improved GPU hardware resilience over A100 GPUs with respect to critical hardware components, (iv) GPU errors on both A100 and H100 GPUs frequently result in job failures due to the lack of robust recovery mechanisms at the application level, and (v) We project the impact of GPU node availability on larger-scales and find that significant overprovisioning of 5% is necessary to handle GPU failures.

IBM (United States) (US), University of Illinois Urbana-Champaign (US), Nokia (Finland) (FI)
Openalex Percentile: Top 22%
Radiation Effects in Electronics
0.73
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Story of Two GPUs: Characterizing the Resilience of Hopper H100 and Ampere A100 GPUs — Phuong Cao, Saurabh Jha, et al. · (2025) | TGRS Research Map | TGRS