dc.creator | Velasco Montero, Delia | es |
dc.creator | Goossens, Bart | es |
dc.creator | Fernández Berni, Jorge | es |
dc.creator | Rodríguez Vázquez, Ángel Benito | es |
dc.creator | Philips, Wilfried | es |
dc.date.accessioned | 2024-02-16T14:47:05Z | |
dc.date.available | 2024-02-16T14:47:05Z | |
dc.date.issued | 2023 | |
dc.identifier.citation | Velasco Montero, D., Goossens, B., Fernández Berni, J., Rodríguez Vázquez, Á.B. y Philips, W. (2023). A Pipelining-Based Heterogeneous Scheduling and Energy-Throughput Optimization Scheme for CNNs Leveraging Apache TVM. IEEE Access, 11, 35007-35021. https://doi.org/10.1109/ACCESS.2023.3264828. | |
dc.identifier.issn | 2169-3536 | es |
dc.identifier.uri | https://hdl.handle.net/11441/155304 | |
dc.description.abstract | Extracting information of interest from continuous video streams is a strongly demanded computer vision task. For the realization of this task at the edge using the current de-facto standard approach, i.e., deep learning, it is critical to optimize key performance metrics such as throughput and energy consumption according to prescribed application requirements. This allows achieving timely decision-making while extending the battery lifetime as much as possible. In this context, we propose a method to boost neural-network performance based on a co-execution strategy that exploits hardware heterogeneity on edge platforms. The enabling tool is Apache TVM, a highly efficient machine-learning compiler compatible with a diversity of hardware back-ends. The proposed approach solves the problem of network partitioning and distributes the workloads to make concurrent use of all the processors available on the board following a pipeline scheme. We conducted experiments on various popular CNNs compiled with TVM on the Jetson TX2 platform. The experimental results based on measurements show a significant improvement in throughput with respect to a single-processor execution, ranging from 14% to 150% over all tested networks. Power-efficient configurations were also identified, accomplishing energy reductions above 10%. | es |
dc.description.sponsorship | Flemish Government PID2021-128009OB-C31 | es |
dc.description.sponsorship | Ministerio de Ciencia e Innovación TED2021-131835B-I00 | es |
dc.description.sponsorship | Office of Naval Research N00014-19-1-2156 | es |
dc.format | application/pdf | es |
dc.format.extent | 15 p. | es |
dc.language.iso | eng | es |
dc.publisher | IEEE | es |
dc.relation.ispartof | IEEE Access, 11, 35007-35021. | |
dc.rights | Attribution-NonCommercial-NoDerivatives 4.0 Internacional | * |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/4.0/ | * |
dc.subject | Apache TVM | es |
dc.subject | Continuous inference | es |
dc.subject | Convolutional neural networks | es |
dc.subject | Edge vision | es |
dc.subject | Heterogeneous processing | es |
dc.subject | Jetson TX2 | es |
dc.subject | Performance optimization | es |
dc.title | A Pipelining-Based Heterogeneous Scheduling and Energy-Throughput Optimization Scheme for CNNs Leveraging Apache TVM | es |
dc.type | info:eu-repo/semantics/article | es |
dc.type.version | info:eu-repo/semantics/publishedVersion | es |
dc.rights.accessRights | info:eu-repo/semantics/openAccess | es |
dc.contributor.affiliation | Universidad de Sevilla. Departamento de Electrónica y Electromagnetismo | es |
dc.relation.projectID | PID2021-128009OB-C31 | es |
dc.relation.projectID | TED2021-131835B-I00 | es |
dc.relation.projectID | N00014-19-1-2156 | es |
dc.relation.publisherversion | https://dx.doi.org/10.1109/ACCESS.2023.3264828 | es |
dc.identifier.doi | 10.1109/ACCESS.2023.3264828 | es |
dc.journaltitle | IEEE Access | es |
dc.publication.volumen | 11 | es |
dc.publication.initialPage | 35007 | es |
dc.publication.endPage | 35021 | es |
dc.contributor.funder | Flemish Government (Belgium) | es |
dc.contributor.funder | Ministerio de Ciencia e Innovación (MICIN). España | es |
dc.contributor.funder | Office of Naval Research (ONR). United States | es |