Matches in SemOpenAlex for { <https://semopenalex.org/work/W3157657667> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W3157657667 abstract "Deep neural networks (DNNs) have been successfully applied to a great variety of applications, ranging from small IoT devices to large scale services in a data center. In order to improve the efficiency of processing these DNN models, dedicated hardware accelerators are required for all these scenarios. Theoretically, there exists an optimized acceleration architecture for each application. However, considering the cost of chip design and corresponding tool-chain development, researchers need to trade off between efficiency and generality. In this work, we demonstrate that it is practical to use a unified architecture, called Ascend, to support those applications, ranging from IoT devices to data-center services. We provide a lot of design details to explain that the success of Ascend relies on contributions from different levels. First, heterogeneous computing units are employed to support various DNN models. And the datapath is adapted according to the requirement of computing and data access. Second, when scaling the Ascend architecture from a single core to a cluster containing thousands of cores, it involves design efforts, such as memory hierarchy and system level integration. Third, a multi-tier compiler, which provides flexible choices for developers, is the last critical piece. Experimental results show that using accelerators based on the Ascend architecture can achieve comparable or even better performance in different applications. In addition, various chips based on the Ascend architecture have been successfully commercialized. More than 100 million chips have been used in real products." @default.
- W3157657667 created "2021-05-10" @default.
- W3157657667 creator A5009275386 @default.
- W3157657667 creator A5030342330 @default.
- W3157657667 creator A5039715250 @default.
- W3157657667 creator A5054361465 @default.
- W3157657667 creator A5054482306 @default.
- W3157657667 creator A5073920552 @default.
- W3157657667 creator A5085132219 @default.
- W3157657667 date "2021-02-01" @default.
- W3157657667 modified "2023-09-30" @default.
- W3157657667 title "Ascend: a Scalable and Unified Architecture for Ubiquitous Deep Neural Network Computing : Industry Track Paper" @default.
- W3157657667 cites W2117130368 @default.
- W3157657667 cites W2161180338 @default.
- W3157657667 cites W2194775991 @default.
- W3157657667 cites W2575705757 @default.
- W3157657667 cites W2585720638 @default.
- W3157657667 cites W2592389822 @default.
- W3157657667 cites W2606722458 @default.
- W3157657667 cites W2734507482 @default.
- W3157657667 cites W2791879367 @default.
- W3157657667 cites W2804686008 @default.
- W3157657667 cites W2889390790 @default.
- W3157657667 cites W2903841424 @default.
- W3157657667 cites W2963087201 @default.
- W3157657667 cites W2966206672 @default.
- W3157657667 cites W2998506323 @default.
- W3157657667 doi "https://doi.org/10.1109/hpca51647.2021.00071" @default.
- W3157657667 hasPublicationYear "2021" @default.
- W3157657667 type Work @default.
- W3157657667 sameAs 3157657667 @default.
- W3157657667 citedByCount "11" @default.
- W3157657667 countsByYear W31576576672022 @default.
- W3157657667 countsByYear W31576576672023 @default.
- W3157657667 crossrefType "proceedings-article" @default.
- W3157657667 hasAuthorship W3157657667A5009275386 @default.
- W3157657667 hasAuthorship W3157657667A5030342330 @default.
- W3157657667 hasAuthorship W3157657667A5039715250 @default.
- W3157657667 hasAuthorship W3157657667A5054361465 @default.
- W3157657667 hasAuthorship W3157657667A5054482306 @default.
- W3157657667 hasAuthorship W3157657667A5073920552 @default.
- W3157657667 hasAuthorship W3157657667A5085132219 @default.
- W3157657667 hasConcept C111919701 @default.
- W3157657667 hasConcept C115051666 @default.
- W3157657667 hasConcept C115537543 @default.
- W3157657667 hasConcept C118524514 @default.
- W3157657667 hasConcept C120314980 @default.
- W3157657667 hasConcept C123657996 @default.
- W3157657667 hasConcept C142362112 @default.
- W3157657667 hasConcept C149635348 @default.
- W3157657667 hasConcept C153349607 @default.
- W3157657667 hasConcept C154240960 @default.
- W3157657667 hasConcept C169590947 @default.
- W3157657667 hasConcept C2777904410 @default.
- W3157657667 hasConcept C2778100165 @default.
- W3157657667 hasConcept C35869016 @default.
- W3157657667 hasConcept C41008148 @default.
- W3157657667 hasConcept C48044578 @default.
- W3157657667 hasConcept C55356503 @default.
- W3157657667 hasConcept C76155785 @default.
- W3157657667 hasConceptScore W3157657667C111919701 @default.
- W3157657667 hasConceptScore W3157657667C115051666 @default.
- W3157657667 hasConceptScore W3157657667C115537543 @default.
- W3157657667 hasConceptScore W3157657667C118524514 @default.
- W3157657667 hasConceptScore W3157657667C120314980 @default.
- W3157657667 hasConceptScore W3157657667C123657996 @default.
- W3157657667 hasConceptScore W3157657667C142362112 @default.
- W3157657667 hasConceptScore W3157657667C149635348 @default.
- W3157657667 hasConceptScore W3157657667C153349607 @default.
- W3157657667 hasConceptScore W3157657667C154240960 @default.
- W3157657667 hasConceptScore W3157657667C169590947 @default.
- W3157657667 hasConceptScore W3157657667C2777904410 @default.
- W3157657667 hasConceptScore W3157657667C2778100165 @default.
- W3157657667 hasConceptScore W3157657667C35869016 @default.
- W3157657667 hasConceptScore W3157657667C41008148 @default.
- W3157657667 hasConceptScore W3157657667C48044578 @default.
- W3157657667 hasConceptScore W3157657667C55356503 @default.
- W3157657667 hasConceptScore W3157657667C76155785 @default.
- W3157657667 hasLocation W31576576671 @default.
- W3157657667 hasOpenAccess W3157657667 @default.
- W3157657667 hasPrimaryLocation W31576576671 @default.
- W3157657667 hasRelatedWork W2048143911 @default.
- W3157657667 hasRelatedWork W2147561140 @default.
- W3157657667 hasRelatedWork W2353633720 @default.
- W3157657667 hasRelatedWork W2364921833 @default.
- W3157657667 hasRelatedWork W2380023786 @default.
- W3157657667 hasRelatedWork W2385146268 @default.
- W3157657667 hasRelatedWork W2390860431 @default.
- W3157657667 hasRelatedWork W2391528269 @default.
- W3157657667 hasRelatedWork W2546696010 @default.
- W3157657667 hasRelatedWork W2511836740 @default.
- W3157657667 isParatext "false" @default.
- W3157657667 isRetracted "false" @default.
- W3157657667 magId "3157657667" @default.
- W3157657667 workType "article" @default.