Repository logo
 
Publication

Implementing CNNs using a linear array of full mesh CGRAs

2020-04Conference object Restricted access
Simple item page
dc.contributor.authorMário, Valter
dc.contributor.authorLopes, João D.
dc.contributor.authorVéstias, Mário
dc.contributor.authorDe Sousa, Jose
dc.date.accessioned2020-04-03T15:24:28Z
dc.date.available2020-04-03T15:24:28Z
dc.date.issued2020-04
dc.description.abstractThis paper presents an implementation of a Convolutional Neural Network (CNN) algorithm using a linear array of full mesh dynamically and partially reconfigurable Coarse Grained Reconfigurable Arrays (CGRAs). Accelerating CNNs using GPUs and FPGAs is more common and there are few works that address the topic of CNN acceleration using CGRAs. Using CGRAs can bring size and power advantages compared to GPUs and FPGAs. The contribution of this paper is to study the performance of full mesh dynamically and partially reconfigurable CGRAs for CNN acceleration. The CGRA used is an improved version of the previously published Versat CGRA, adding multi CGRA core support and pre-silicon configurability. The results show that the proposed CGRA is as easy to program as the original full mesh Versat CGRA, and that its performance and power consumption scale linearly with the number of instances.pt_PT
dc.description.versioninfo:eu-repo/semantics/publishedVersionpt_PT
dc.identifier.citationMÁRIO, Valter; [et al] – Implementing CNNs using a linear array of full mesh CGRAs. In Applied Reconfigurable Computing – Proceedings of the 16th International Symposium on Applied Reconfigurable Computing. Toledo, Spain: Springer, 2020. Vol. 12083, pp. 1-10pt_PT
dc.identifier.urihttp://hdl.handle.net/10400.21/11401
dc.language.isoengpt_PT
dc.publisherSpringerpt_PT
dc.relationPTDC/EEI-HAC/30848/2017 - FCTpt_PT
dc.relationUIDB/50021/2020 - FCTpt_PT
dc.subjectConvolutional Neural Networkspt_PT
dc.subjectCoarse Grained Reconfigurable Arrayspt_PT
dc.subjectReconfigurable computingpt_PT
dc.subjectEmbedded systemspt_PT
dc.titleImplementing CNNs using a linear array of full mesh CGRAspt_PT
dc.typeconference object
dspace.entity.typePublication
oaire.citation.conferencePlaceApril 1–3, 2020 – Toledo, Spainpt_PT
oaire.citation.endPage10pt_PT
oaire.citation.startPage1pt_PT
oaire.citation.title16th International Symposium on Applied Reconfigurable Computingpt_PT
oaire.citation.volume12083pt_PT
person.familyNameVéstias
person.familyNamede Sousa
person.givenNameMário
person.givenNameJose
person.identifier.ciencia-id4717-C2C7-3F2C
person.identifier.ciencia-idBE18-E262-E0EC
person.identifier.orcid0000-0001-8556-4507
person.identifier.orcid0000-0001-7525-7546
person.identifier.ridH-9953-2012
person.identifier.ridL-6859-2015
person.identifier.scopus-author-id14525867300
person.identifier.scopus-author-id7102813024
rcaap.rightsclosedAccesspt_PT
rcaap.typeconferenceObjectpt_PT
relation.isAuthorOfPublicationa7d22b29-c961-45ac-bc09-cd5e1002f1e8
relation.isAuthorOfPublicationd98a4d45-2d45-42ec-9f1d-14775723709b
relation.isAuthorOfPublication.latestForDiscoverya7d22b29-c961-45ac-bc09-cd5e1002f1e8

Files

Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
Implementing_MVestias.pdf
Size:
463.64 KB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

ISEL - Eng. Elect. Tel. Comp. - Comunicações