SASSIFI: An architecture-level fault injection tool for GPU application resilience evaluation
The DOI matches its Crossref record; 24 of the 24 references it cites were found in a registry or an index.
Abstract
As GPUs become more pervasive in both scalable high-performance computing systems and safety-critical embedded systems, evaluating and analyzing their resilience to soft errors caused by high-energy particle strikes will grow increasingly important. GPU designers must develop tools and techniques to understand the effect of these soft errors on applications. This paper presents an error injection-based methodology and tool called SASSIFI to study the soft error resilience of massively parallel applications running on state-of-the-art NVIDIA GPUs. Our approach uses a low-level assembly-language instrumentation tool called SASSI to profile and inject errors. SASSI provides efficiency by allowing instrumentation code to execute entirely on the GPU and provides the ability to inject into different architecture-visible state. For example, SASSIFI can inject errors in general-purpose registers, GPU memory, condition code registers, and predicate registers. SASSIFI can also inject errors into addresses and register indices. In this paper, we describe the SASSIFI tool, its capabilities, and present experiments to illustrate some of the analyses SASSIFI can be used to perform.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Architecture-Level Fault Injection , FI in short, provides controllability and observability, as well as high accuracy, because of the ability of emulating the physical effects of the faults in the architecture components.”
From Fast and Accurate Error Simulation for CNNs Against Soft Errors (2023) · cites this paper as [15]
In the citing paper
“When considering the GPU platform, as discussed in [15], an SEU occurring in the SM may corrupt the execution of a single thread causing an SDC, or a group of threads within the same warp belonging to a single kernel; moreover, when shared memory is used, erroneous data produced by a thread may propagate among several threads within the same block of the kernel.”
From Fast and Accurate Error Simulation for CNNs Against Soft Errors (2023) · cites this paper as [15]
Citations
Cites
Rodinia: A benchmark suite for heterogeneous computing
Che, S., Boyer, M., Meng, J., Tarjan, D., Sheaffer, J. W., Lee, S.-H., & Skadron, K. (2009) · 2009 IEEE International Symposium on Workload Characterization (IISWC) · doi:10.1109/iiswc.2009.5306797 · Cited by 3,018
Mukherjee, S. S., Weaver, C. T., Emer, J., Reinhardt, S. K., & Austin, T. M. (2003) · Cited by 840
Self-citationMukherjee, S. S., Weaver, C., Emer, J., Reinhardt, S. K., & Austin, T. (2004) · 22nd Digital Avionics Systems Conference. Proceedings (Cat. No.03CH37449) · doi:10.1109/micro.2003.1253181 · Cited by 656
Self-citationQuantitative evaluation of soft error injection techniques for robust system design
Cho, H., Mirkhani, S., Cher, C.-Y., Abraham, J. A., & Mitra, S. (2013) · Proceedings of the 50th Annual Design Automation Conference · doi:10.1145/2463209.2488859 · Cited by 215
Celio, C., Patterson, D. A., & Asanović, K. (2015) · UC Berkeley · Cited by 210
Computing Architectural Vulnerability Factors for Address-Based Structures
Biswas, A., Racunas, P., Cheveresan, R., Emer, J., Mukherjee, S. S., & Rangan, R. (2005) · ACM SIGARCH Computer Architecture News · doi:10.1145/1080695.1070014 · Cited by 184
Self-citationComputing Architectural Vulnerability Factors for Address-Based Structures
Biswas, A., Racunas, P., Cheveresan, R., Emer, J., Mukherjee, S. S., & Rangan, R. (2005) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1109/isca.2005.18 · Cited by 183
Self-citationQuantifying the Accuracy of High-Level Fault Injection Techniques for Hardware Faults
Wei, J., Thomas, A., Li, G., & Pattabiraman, K. (2014) · 2014 44th Annual IEEE/IFIP International Conference on Dependable Systems and Networks · doi:10.1109/dsn.2014.2 · Cited by 182 · cited 1×
“2) What is the probability that a bit-flip in the destination register of an executing instruction will result in an SDC? This study aims to quantify the effect of bit-flips in low-level unprotected state at the application-level by injecting the manifestation at the architecture-level, a commonly studied topic [6], […”
BackgroundEliminating microarchitectural dependency from Architectural Vulnerability
Sridharan, V., & Kaeli, D. R. (2009) · 2009 IEEE 15th International Symposium on High Performance Computer Architecture · doi:10.1109/hpca.2009.4798243 · Cited by 182
An Experimental Study of Soft Errors in Microprocessors
Saggese, G. P., Wang, N. J., Kalbarczyk, Z. T., Patel, S. J., & Iyer, R. K. (2005) · IEEE Micro · doi:10.1109/mm.2005.104 · Cited by 160
Research works printed but not linked: 9
- 2015, Profile User's Guide
- 2012, NVIDIA's Next Generation CUDA Compute Architecture Kepler GK110
- 2014, NVIDIA GeForce GTX 980 Featuring Maxwell The Most Advanced GPU Ever Made
- 2014, CUDA-GDB CUDA Toolkit Documentation
- 2015, PTX ISA CUDA Toolkit Documentation
- 2013, Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics (LULESH) 2.0, Lawrence Livermore National Laboratory
- 0, CoMD: Classical Molecular Dynamics Proxy Application, Exascale Co-Design Center for Materials in Extreme Environments (ExMatEx)
- 2014, CUPTI: : CUDA Toolkit Documentation, Nvidia
- farazmand, 2012, Statistical Fault Injection-Based AVF Analysis of a GPU Architecure, Proc 2nd Workshop Silicon Errors Logic-Syst Effects (SELSE)
Cited by
Items 201–219 of 219
ResCiM: Exploring Reliability of Hybrid Floating-point Compute-in-Memory Deep Learning Accelerators
Bhattacharya, S., Das, S., Menon, A., Kundu, S., Raha, A., & Basu, K. (2026) · IEEE transactions on circuits and systems for artificial intelligence. · doi:10.1109/tcasai.2026.3699496
Alpay, F., & Basaran, B. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.21791
Beyond Price: A Technical Quality Framework for AI Antitrust
Wu, P. Y. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812338
Resilience Assessment of AI Accelerators: A Driver-Level Fault Injection Methodology
Cinque, M., Simone, L., & Giacomo, N. D. (2026) · International Conference on Dependable Systems and Networks workshops · doi:10.1109/dsn-w70714.2026.00042
Dennis, W., & Pope, J. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-25032-2_10
Yoshida, M., Matsuoka, K., & Hashimoto, M. (2026) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2026.3714766
A GraphSAGE-Based Register-Level Soft Error Rate Prediction Framework for Selective Hardening
Lu, J., Lin, Y., Hu, Y., Bao, K., Xu, C., & Liu, Y. (2026) · IEEE International Conference on Electronics, Communications and Intelligent Science (ECIS) · doi:10.1109/ecis69634.2026.11604281
Adaptive Fault Injection Planning for Multi-Layer Self-Healing AI InfrastructurePreprint
Kulkarni, S. R., Yang, Y., Kulkarni, R., & Nayak, G. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.16161
CIM-FI: A HW-Aware Fault Injection Framework for Digital Compute-In-Memory DNN Accelerators
Chaidos, P., Maras, A., Spyrou, T., Gebregiorgis, A., Hamdioui, S., Soudris, D., & Xydis, S. (2026) · IEEE International On-Line Testing Symposium (IOLTS) · doi:10.1109/iolts69666.2026.11633664
Mind the Driver: A Fault Injection Methodology for AI Systems
Cinque, M., Simone, L., & Giacomo, N. D. (2026) · IEEE/IFIP International Conference on Dependable Systems and Networks - Supplemental Volume (DSN-S) · doi:10.1109/dsn-s70715.2026.00076
Lu, J., Xu, C., Huixin, P., Zhang, G., Wang, L., Liao, X., Liu, Y., & Yang, Y. (2026) · Micromachines · doi:10.3390/mi17091059
Scalable fault vulnerability analysis for ReRAM-based process-in-memory CNN accelerators
Dorostkar, A., Farbeh, H., Zarandi, H. R., & Sharifzadeh, M. (2026) · The Journal of Supercomputing · doi:10.1007/s11227-026-08867-x
New techniques for the Reliability Evaluation of AI-oriented Hardware Accelerators
Sierra, R. L. (2026) · PORTO@Iris
Reliability Enhancement in GPU Architectures
Guerrero-Balaguera, J.-D. (2024) · PORTO@Iris
BitIR: Cross-Architecture Fault Injection for Resilience Analysis of Heterogeneous GPU ApplicationsPreprint
Dunlavy, M., Papka, M., & Lan, Z. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.04037
When Log Parsing Removes What Matters: Detecting Failures in Distributed GPU Workloads
Cui, B., Zeng, Y., Tian, Y., & Jacobsen, H. (2026) · doi:10.1145/3820756.3844919
DFIT: A Framework for Building Instrumentation Tools Across Heterogeneous Dataflow AcceleratorsPreprint
Alam, M. A., Hossain, A., & Ahmed, K. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.14280
Zou, F., Wang, B., Qiu, Y., Wang, C., & Wang, Z. (2026) · International Symposium of Electronics Design Automation (ISEDA) · doi:10.1109/iseda70083.2026.11751440
From Tool State to Scientific Action: Auditing Decision Provenance in Biomedical Research AgentsPreprint
Yi, X., Zhang, X., Zhou, Y., & Bao, H. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2611.01527
Metadata
- Authors
- Hari, S. K. S., Tsai, T., Stephenson, M., Keckler, S. W., & Emer, J.
- Affiliations
- Nvidia (United Kingdom) (GB)
- Venue
- 2017 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)
- Publisher
- IEEE
- Published
- 2017-04
- Pages
- 249-258
- OpenAlex
- W2735162286
- Cited by
- 181
- References
- 24
- Citations by year
- 2026: 15 · 2025: 11 · 2024: 13 · 2023: 24 · 2022: 22 · 2021: 22 · 2020: 24 · 2019: 23 · 2018: 22 · 2017: 5
- FWCI
- 28.63
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS) (publishedVersion)
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Radiation Effects in Electronics
- Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
- Physical Sciences › Computer Science › Computer Networks and Communications › Distributed systems and fault tolerance
- Keywords
- fault injection, soft errors, SASSI, NVIDIA GPUs, register errors, memory errors, safety-critical embedded systems, high-performance computing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.