Matches in SemOpenAlex for { <https://semopenalex.org/work/W2087031469> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W2087031469 endingPage "65" @default.
- W2087031469 startingPage "50" @default.
- W2087031469 abstract "We improve the dual modular redundancy (DMR), a soft error detection scheme.We propose a grid sampling error detection scheme for stencil-based computation.We analyze the soft error propagation pattern through the stencil operations.We demonstrate our scheme by experiments on Tianhe-2 supercomputer. Soft errors are becoming a prominent problem for massive parallel scientific applications. Dual-modular redundancy (DMR) can provide approximately 100% error coverage, but it has the problem of overhead excessive. Stencil kernel is one of the most important routines applied in the context of structured grids. In this paper, we propose Grid Sampling DMR (GS-DMR), a low-overhead soft error detection scheme for stencil-based computation. Instead of comparing the whole set of the results in the traditional DMR, GS-DMR just compares a subset of the results according to sampling on the grid data, which is based on the error propagation pattern on the grid. We also design a fault tolerant (FT) framework combining GS-DMR with checkpoint technology, and provide theoretical analysis and an algorithm for the optimal FT parameters. Experimental results on Tianhe-2 supercomputer demonstrate that GS-DMR can achieve a good FT effect for stencil-based computation, and the effect is greatly improved for massively parallel applications, reducing the total FT overhead up to 51%." @default.
- W2087031469 created "2016-06-24" @default.
- W2087031469 creator A5004004713 @default.
- W2087031469 creator A5016781380 @default.
- W2087031469 creator A5026071162 @default.
- W2087031469 creator A5055062652 @default.
- W2087031469 creator A5059956624 @default.
- W2087031469 creator A5061870286 @default.
- W2087031469 date "2015-01-01" @default.
- W2087031469 modified "2023-10-16" @default.
- W2087031469 title "GS-DMR: Low-overhead soft error detection scheme for stencil-based computation" @default.
- W2087031469 cites W2001219263 @default.
- W2087031469 cites W2031533321 @default.
- W2087031469 cites W2033656974 @default.
- W2087031469 cites W2039345646 @default.
- W2087031469 cites W2041028442 @default.
- W2087031469 cites W2047360090 @default.
- W2087031469 cites W2049472804 @default.
- W2087031469 cites W2052455844 @default.
- W2087031469 cites W2083613288 @default.
- W2087031469 cites W2102576651 @default.
- W2087031469 cites W2103068098 @default.
- W2087031469 cites W2133046454 @default.
- W2087031469 cites W2136905849 @default.
- W2087031469 cites W2152465173 @default.
- W2087031469 cites W2169624732 @default.
- W2087031469 cites W2174151054 @default.
- W2087031469 cites W3111197091 @default.
- W2087031469 doi "https://doi.org/10.1016/j.parco.2014.11.003" @default.
- W2087031469 hasPublicationYear "2015" @default.
- W2087031469 type Work @default.
- W2087031469 sameAs 2087031469 @default.
- W2087031469 citedByCount "10" @default.
- W2087031469 countsByYear W20870314692015 @default.
- W2087031469 countsByYear W20870314692016 @default.
- W2087031469 countsByYear W20870314692017 @default.
- W2087031469 countsByYear W20870314692018 @default.
- W2087031469 countsByYear W20870314692020 @default.
- W2087031469 countsByYear W20870314692022 @default.
- W2087031469 crossrefType "journal-article" @default.
- W2087031469 hasAuthorship W2087031469A5004004713 @default.
- W2087031469 hasAuthorship W2087031469A5016781380 @default.
- W2087031469 hasAuthorship W2087031469A5026071162 @default.
- W2087031469 hasAuthorship W2087031469A5055062652 @default.
- W2087031469 hasAuthorship W2087031469A5059956624 @default.
- W2087031469 hasAuthorship W2087031469A5061870286 @default.
- W2087031469 hasConcept C111919701 @default.
- W2087031469 hasConcept C11413529 @default.
- W2087031469 hasConcept C127413603 @default.
- W2087031469 hasConcept C134306372 @default.
- W2087031469 hasConcept C154474529 @default.
- W2087031469 hasConcept C173608175 @default.
- W2087031469 hasConcept C24326235 @default.
- W2087031469 hasConcept C2779960059 @default.
- W2087031469 hasConcept C33923547 @default.
- W2087031469 hasConcept C41008148 @default.
- W2087031469 hasConcept C45374587 @default.
- W2087031469 hasConcept C459310 @default.
- W2087031469 hasConcept C76752949 @default.
- W2087031469 hasConcept C77618280 @default.
- W2087031469 hasConceptScore W2087031469C111919701 @default.
- W2087031469 hasConceptScore W2087031469C11413529 @default.
- W2087031469 hasConceptScore W2087031469C127413603 @default.
- W2087031469 hasConceptScore W2087031469C134306372 @default.
- W2087031469 hasConceptScore W2087031469C154474529 @default.
- W2087031469 hasConceptScore W2087031469C173608175 @default.
- W2087031469 hasConceptScore W2087031469C24326235 @default.
- W2087031469 hasConceptScore W2087031469C2779960059 @default.
- W2087031469 hasConceptScore W2087031469C33923547 @default.
- W2087031469 hasConceptScore W2087031469C41008148 @default.
- W2087031469 hasConceptScore W2087031469C45374587 @default.
- W2087031469 hasConceptScore W2087031469C459310 @default.
- W2087031469 hasConceptScore W2087031469C76752949 @default.
- W2087031469 hasConceptScore W2087031469C77618280 @default.
- W2087031469 hasFunder F4320321001 @default.
- W2087031469 hasLocation W20870314691 @default.
- W2087031469 hasOpenAccess W2087031469 @default.
- W2087031469 hasPrimaryLocation W20870314691 @default.
- W2087031469 hasRelatedWork W1516738056 @default.
- W2087031469 hasRelatedWork W2035701183 @default.
- W2087031469 hasRelatedWork W2083051666 @default.
- W2087031469 hasRelatedWork W2149668662 @default.
- W2087031469 hasRelatedWork W2497986104 @default.
- W2087031469 hasRelatedWork W2804132206 @default.
- W2087031469 hasRelatedWork W3091752332 @default.
- W2087031469 hasRelatedWork W3122909934 @default.
- W2087031469 hasRelatedWork W4200123712 @default.
- W2087031469 hasRelatedWork W4310509441 @default.
- W2087031469 hasVolume "41" @default.
- W2087031469 isParatext "false" @default.
- W2087031469 isRetracted "false" @default.
- W2087031469 magId "2087031469" @default.
- W2087031469 workType "article" @default.