{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T14:00:05Z","timestamp":1775743205280,"version":"3.50.1"},"reference-count":15,"publisher":"Springer Science and Business Media LLC","issue":"5","license":[{"start":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T00:00:00Z","timestamp":1775692800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T00:00:00Z","timestamp":1775692800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Supercomput"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Mixed-precision arithmetic, widely adopted for performance, creates an overlooked opportunity for fault tolerance. Our mixed-precision panel factorization () algorithm already computes each LU panel in both  and  ; we show that this inherent redundancy enables soft-error detection at negligible additional cost. We propose four complementary detectors that compare checksum sketches of the two factorizations:  (thresholded checksum discrepancy),  (outlier analysis of per-row differences),  (a randomized-probe combination of both with separate\n                    <jats:inline-formula>\n                      <jats:alternatives>\n                        <jats:tex-math>$$L$$<\/jats:tex-math>\n                        <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                          <mml:mi>L<\/mml:mi>\n                        <\/mml:math>\n                      <\/jats:alternatives>\n                    <\/jats:inline-formula>\n                    - and\n                    <jats:inline-formula>\n                      <jats:alternatives>\n                        <jats:tex-math>$$U$$<\/jats:tex-math>\n                        <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                          <mml:mi>U<\/mml:mi>\n                        <\/mml:math>\n                      <\/jats:alternatives>\n                    <\/jats:inline-formula>\n                    -factor checks), and  (a lightweight cross-check inspired by algorithm-based fault tolerance). Experimental evaluation shows that  and  achieve high sensitivity with low false-positive rates, while  maintains near-zero false positives with moderate sensitivity and complementary coverage for challenging significand-bit faults. Bit-level analysis confirms reliable detection of exponent- and sign-bit errors, with sensitivity decreasing gracefully for low-order significand bits whose perturbations approach the  rounding floor. A lightweight panel-fingerprint mechanism extends protection beyond the factorization loop, closing the temporal gap before the triangular solve with guaranteed detection of any single-element corruption and zero false positives. The approach requires only linear work and constant storage per panel, preserving the cubic scaling of standard LU decomposition.\n                  <\/jats:p>","DOI":"10.1007\/s11227-026-08413-9","type":"journal-article","created":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T13:07:20Z","timestamp":1775740040000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Exploiting mixed-precision redundancy for soft-error detection in LU decomposition"],"prefix":"10.1007","volume":"82","author":[{"given":"Nima","family":"Sahraneshinsamani","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sandra","family":"Catal\u00e1n","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jos\u00e9 R.","family":"Herrero","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2026,4,9]]},"reference":[{"issue":"1","key":"8413_CR1","doi-asserted-by":"publisher","first-page":"87","DOI":"10.1007\/s11227-024-06523-w","volume":"81","author":"N Sahraneshinsamani","year":"2024","unstructured":"Sahraneshinsamani N, Catal\u00e1n S, Herrero JR (2024) Mixed-precision pre-pivoting strategy for the LU factorization. J Supercomput 81(1):87. https:\/\/doi.org\/10.1007\/s11227-024-06523-w","journal-title":"J Supercomput"},{"issue":"4","key":"8413_CR2","doi-asserted-by":"publisher","first-page":"337","DOI":"10.1109\/TDSC.2009.4","volume":"7","author":"B Schroeder","year":"2010","unstructured":"Schroeder B, Gibson GA (2010) A large-scale study of failures in high-performance computing systems. IEEE Trans Dependable Secure Comput 7(4):337\u2013350. https:\/\/doi.org\/10.1109\/TDSC.2009.4","journal-title":"IEEE Trans Dependable Secure Comput"},{"key":"8413_CR3","doi-asserted-by":"publisher","unstructured":"Davies T, Chen Z (2013) Correcting soft errors online in LU factorization. In: Proceedings of the 22nd International Symposium on High-performance Parallel and Distributed Computing (HPDC \u201913). Association for Computing Machinery, New York, NY, USA, pp. 167\u2013178 https:\/\/doi.org\/10.1145\/2462902.2462920","DOI":"10.1145\/2462902.2462920"},{"key":"8413_CR4","doi-asserted-by":"publisher","unstructured":"Haidar A, Tomov S, Dongarra J, Higham NJ (2018) Harnessing GPU tensor cores for fast FP16 arithmetic to speed up mixed-precision iterative refinement solvers. In: SC \u201918: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis. IEEE Press, Dallas, TX, USA, pp. 47\u201314711 https:\/\/doi.org\/10.1145\/3291656.3291719","DOI":"10.1145\/3291656.3291719"},{"key":"8413_CR5","doi-asserted-by":"publisher","first-page":"347","DOI":"10.1017\/S0962492922000022","volume":"31","author":"NJ Higham","year":"2022","unstructured":"Higham NJ, Mary T (2022) Mixed-precision algorithms in numerical linear algebra. Acta Numer 31:347\u2013414. https:\/\/doi.org\/10.1017\/S0962492922000022","journal-title":"Acta Numer"},{"issue":"9","key":"8413_CR6","doi-asserted-by":"publisher","first-page":"12844","DOI":"10.1007\/s11227-024-05925-0","volume":"80","author":"G Leon","year":"2024","unstructured":"Leon G, Badia JM, Belloch JA, Lindoso A, Entrena L (2024) Comparative analysis of soft-error sensitivity in LU decomposition algorithms on diverse GPUs. J Supercomput 80(9):12844\u201312862. https:\/\/doi.org\/10.1007\/s11227-024-05925-0","journal-title":"J Supercomput"},{"issue":"12","key":"8413_CR7","first-page":"4424","volume":"33","author":"J Fang","year":"2022","unstructured":"Fang J, Varbanescu AL, Imbernon B, Brunelle P, Peltenburg J, Sips H, Gai H, Purnaprajna M, Werkhoven B (2022) Towards precision-aware fault tolerance approaches for mixed-precision applications on GPUs. IEEE Trans Parallel Distrib Syst 33(12):4424\u20134439","journal-title":"IEEE Trans Parallel Distrib Syst"},{"key":"8413_CR8","unstructured":"Chishti Z, Alameldeen AR, Wilkerson C, Wu W, Lu S-L (2022) Characterizing and mitigating soft errors in GPU DRAM. In: 2022 IEEE International Symposium on High-Performance Computer Architecture (HPCA), pp. 688\u2013701"},{"issue":"6","key":"8413_CR9","doi-asserted-by":"publisher","first-page":"518","DOI":"10.1109\/TC.1984.1676475","volume":"C\u201333","author":"K-H Huang","year":"1984","unstructured":"Huang K-H, Abraham JA (1984) Algorithm-based fault tolerance for matrix operations. IEEE Trans Comput C\u201333(6):518\u2013528. https:\/\/doi.org\/10.1109\/TC.1984.1676475","journal-title":"IEEE Trans Comput"},{"issue":"10","key":"8413_CR10","first-page":"16","volume":"49","author":"J Dongarra","year":"2016","unstructured":"Dongarra J, H\u00e9rault T, Robert Y (2016) Fault tolerance techniques for high-performance computing. Computer 49(10):16\u201325","journal-title":"Computer"},{"key":"8413_CR11","unstructured":"Tremodeux A, Agullo E, Benoit A, Giraud L, H\u00e9rault T, Robert Y (2025) Fault-tolerant numerical iterative algorithms at scale. Technical Report RR-9567, Inria Lyon. https:\/\/inria.hal.science\/hal-04872041"},{"key":"8413_CR12","doi-asserted-by":"publisher","DOI":"10.1137\/1.9780898718027","volume-title":"Accuracy and stability of numerical algorithms","author":"NJ Higham","year":"2002","unstructured":"Higham NJ (2002) Accuracy and stability of numerical algorithms, 2nd edn. Society for Industrial and Applied Mathematics, Philadelphia, PA","edition":"2"},{"key":"8413_CR13","doi-asserted-by":"publisher","unstructured":"Tam N, Bhuva BL, Massengill LW, Ball D, McCurdy M, Alles ML, Chatterjee I (2015) Multi-cell soft errors at the 16-nm FinFET technology node. In: 2015 IEEE International Reliability Physics Symposium, pp. 4\u201331435. https:\/\/doi.org\/10.1109\/IRPS.2015.7112730","DOI":"10.1109\/IRPS.2015.7112730"},{"issue":"16","key":"8413_CR14","doi-asserted-by":"publisher","first-page":"5356","DOI":"10.3390\/s24165356","volume":"24","author":"W Wang","year":"2024","unstructured":"Wang W, Li X, Chen L, Sun H, Zhang F (2024) A review on soft error correcting techniques of aerospace-grade static RAM-based field-programmable gate arrays. Sensors 24(16):5356. https:\/\/doi.org\/10.3390\/s24165356","journal-title":"Sensors"},{"issue":"158","key":"8413_CR15","doi-asserted-by":"publisher","first-page":"209","DOI":"10.1080\/01621459.1927.10502953","volume":"22","author":"EB Wilson","year":"1927","unstructured":"Wilson EB (1927) Probable inference, the law of succession, and statistical inference. J Am Stat Assoc 22(158):209\u2013212","journal-title":"J Am Stat Assoc"}],"container-title":["The Journal of Supercomputing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11227-026-08413-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s11227-026-08413-9","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11227-026-08413-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T13:07:22Z","timestamp":1775740042000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s11227-026-08413-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,9]]},"references-count":15,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,4]]}},"alternative-id":["8413"],"URL":"https:\/\/doi.org\/10.1007\/s11227-026-08413-9","relation":{},"ISSN":["1573-0484"],"issn-type":[{"value":"1573-0484","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,9]]},"assertion":[{"value":"15 October 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"1 March 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"9 April 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The implementation of the MPF algorithm with fault detection is available at\n                      \n                      .","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Code availability"}},{"value":"The authors declare no conflict of interest.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}],"article-number":"336"}}