Matches in SemOpenAlex for { <https://semopenalex.org/work/W4300450771> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4300450771 abstract "Process variations and device aging impose profound challenges for circuit designers. Without a precise understanding of the impact of variations on the delay of circuit paths, guardbands, which keep timing violations at bay, cannot be correctly estimated. This problem is exacerbated for advanced technology nodes, where transistor dimensions reach atomic levels and established margins are severely constrained. Hence, traditional worst-case analysis becomes impractical, resulting in intolerable performance overheads. Contrarily, process-variation/aging-aware static timing analysis (STA) equips designers with accurate statistical delay distributions. Timing guardbands that are small, yet sufficient, can then be effectively estimated. However, such analysis is costly as it requires intensive Monte-Carlo simulations. Further, it necessitates access to confidential physics-based aging models to generate the standard-cell libraries required for STA. In this work, we employ graph neural networks (GNNs) to accurately estimate the impact of process variations and device aging on the delay of any path within a circuit. Our proposed GNN4REL framework empowers designers to perform rapid and accurate reliability estimations without accessing transistor models, standard-cell libraries, or even STA; these components are all incorporated into the GNN model via training by the foundry. Specifically, GNN4REL is trained on a FinFET technology model that is calibrated against industrial 14nm measurement data. Through our extensive experiments on EPFL and ITC-99 benchmarks, as well as RISC-V processors, we successfully estimate delay degradations of all paths -- notably within seconds -- with a mean absolute error down to 0.01 percentage points." @default.
- W4300450771 created "2022-10-03" @default.
- W4300450771 creator A5030348801 @default.
- W4300450771 creator A5052751303 @default.
- W4300450771 creator A5059987567 @default.
- W4300450771 creator A5083905801 @default.
- W4300450771 creator A5087244134 @default.
- W4300450771 date "2022-08-04" @default.
- W4300450771 modified "2023-10-16" @default.
- W4300450771 title "GNN4REL: Graph Neural Networks for Predicting Circuit Reliability Degradation" @default.
- W4300450771 doi "https://doi.org/10.48550/arxiv.2208.02868" @default.
- W4300450771 hasPublicationYear "2022" @default.
- W4300450771 type Work @default.
- W4300450771 citedByCount "1" @default.
- W4300450771 countsByYear W43004507712022 @default.
- W4300450771 crossrefType "posted-content" @default.
- W4300450771 hasAuthorship W4300450771A5030348801 @default.
- W4300450771 hasAuthorship W4300450771A5052751303 @default.
- W4300450771 hasAuthorship W4300450771A5059987567 @default.
- W4300450771 hasAuthorship W4300450771A5083905801 @default.
- W4300450771 hasAuthorship W4300450771A5087244134 @default.
- W4300450771 hasBestOaLocation W43004507711 @default.
- W4300450771 hasConcept C105795698 @default.
- W4300450771 hasConcept C111919701 @default.
- W4300450771 hasConcept C113775141 @default.
- W4300450771 hasConcept C119599485 @default.
- W4300450771 hasConcept C121332964 @default.
- W4300450771 hasConcept C127413603 @default.
- W4300450771 hasConcept C149635348 @default.
- W4300450771 hasConcept C163258240 @default.
- W4300450771 hasConcept C165801399 @default.
- W4300450771 hasConcept C172385210 @default.
- W4300450771 hasConcept C19499675 @default.
- W4300450771 hasConcept C200601418 @default.
- W4300450771 hasConcept C33923547 @default.
- W4300450771 hasConcept C41008148 @default.
- W4300450771 hasConcept C43214815 @default.
- W4300450771 hasConcept C62520636 @default.
- W4300450771 hasConcept C93389723 @default.
- W4300450771 hasConcept C93682380 @default.
- W4300450771 hasConcept C98045186 @default.
- W4300450771 hasConceptScore W4300450771C105795698 @default.
- W4300450771 hasConceptScore W4300450771C111919701 @default.
- W4300450771 hasConceptScore W4300450771C113775141 @default.
- W4300450771 hasConceptScore W4300450771C119599485 @default.
- W4300450771 hasConceptScore W4300450771C121332964 @default.
- W4300450771 hasConceptScore W4300450771C127413603 @default.
- W4300450771 hasConceptScore W4300450771C149635348 @default.
- W4300450771 hasConceptScore W4300450771C163258240 @default.
- W4300450771 hasConceptScore W4300450771C165801399 @default.
- W4300450771 hasConceptScore W4300450771C172385210 @default.
- W4300450771 hasConceptScore W4300450771C19499675 @default.
- W4300450771 hasConceptScore W4300450771C200601418 @default.
- W4300450771 hasConceptScore W4300450771C33923547 @default.
- W4300450771 hasConceptScore W4300450771C41008148 @default.
- W4300450771 hasConceptScore W4300450771C43214815 @default.
- W4300450771 hasConceptScore W4300450771C62520636 @default.
- W4300450771 hasConceptScore W4300450771C93389723 @default.
- W4300450771 hasConceptScore W4300450771C93682380 @default.
- W4300450771 hasConceptScore W4300450771C98045186 @default.
- W4300450771 hasLocation W43004507711 @default.
- W4300450771 hasOpenAccess W4300450771 @default.
- W4300450771 hasPrimaryLocation W43004507711 @default.
- W4300450771 hasRelatedWork W1486460655 @default.
- W4300450771 hasRelatedWork W1995024251 @default.
- W4300450771 hasRelatedWork W2091532617 @default.
- W4300450771 hasRelatedWork W2167870875 @default.
- W4300450771 hasRelatedWork W2296498951 @default.
- W4300450771 hasRelatedWork W2414679839 @default.
- W4300450771 hasRelatedWork W2921149048 @default.
- W4300450771 hasRelatedWork W4206024035 @default.
- W4300450771 hasRelatedWork W4220664679 @default.
- W4300450771 hasRelatedWork W2087720490 @default.
- W4300450771 isParatext "false" @default.
- W4300450771 isRetracted "false" @default.
- W4300450771 workType "article" @default.