Show simple item record

dc.contributor.authorPascual Saiz, José Antonio ORCID
dc.contributor.authorNavaridas Palma, Javier
dc.date.accessioned2020-01-17T19:27:58Z
dc.date.available2020-01-17T19:27:58Z
dc.date.issued2019
dc.identifier.citationEkaia 35 : 309-324 (2019)
dc.identifier.issn0214-9001
dc.identifier.urihttp://hdl.handle.net/10810/39010
dc.description.abstractExascale performance will be delivered by systems composed of millions of inter-connected computing cores. The way these computing elements are connected with each other (network topology) has a strong impact on many performance characteristics. In this work we propose a multi-objective optimization- based framework to explore possible network topologies to be implemented in the EU-funded ExaNeSt project. The modular design of this system’s inter-connect provides great flexibility to design topologies optimized for specific performance targets such as communications locality, fault tolerance or energy-consumption. The generation proce-dure of the topologies is formulated as a three-objective optimization problem (minimizing some topological characteristics) where solutions are searched using evolutionary techniques. The analysis of the results, carried out using simulation, shows that the topologies meet the required performance objectives. In addition, a comparison with a well-known topology reveals that the generated solutions can provide better topological characteristics and also higher throughput in almost all evaluated scenarios.; Exaeskala errendimendua milioika kalkulu-nukleoz osaturiko sistemak erabi-liz lortuko da. Elementu horiek konektatzeko moduak (sare-topologia) izugarrizko eragina du hainbat errendimendu ezaugarritan. Lan honetan, sare-topologiak diseinatzeko helburu-anizta-suneko optimizazioan oinarritutako ingurune bat proposatzen dugu, EBk finantzatuta ExaNeSt proiektuan garatzen ari garena. ExaNeSt sistemaren sarearen modulartasunari esker sare-topolo-gia ezberdinak diseinatu ditzakegu hainbat errendimendu-helburu optimizatzeko; esaterako, in-guruko komunikazioak, hutsegite-tolerantzia eta energia-kontsumoa. Topologiak sortzeko pro-zesua optimizazio bidez gauzatzen da (sare-topologiaren hainbat ezaugarri minimizatuz) teknika ebolutiboak erabilita. Simulazio bidezko emaitzen analisiak sortutako topologiek errendimen-du-helburuak betetzen dituztela erakusten du. Gainera, sare-topologia ezagun batekin egindako konparazioan ikus daiteke gure proposamenak sortzen dituen sareek propietate topologiko ho-beak dauzkatela eta, aldi berean, errendimendu handiagoa lortzen dutela.
dc.language.isoeus
dc.publisherServicio Editorial de la Universidad del País Vasco/Euskal Herriko Unibertsitatearen Argitalpen Zerbitzua
dc.rightsinfo:eu-repo/semantics/openAccess
dc.rights.urihttp://creativecommons.org/licenses/by-nc-sa/4.0/
dc.titleExaeskalarako sare-interkonexioen diseinurako helburu-aniztasuneko optimizazioa
dc.typeinfo:eu-repo/semantics/article
dc.rights.holder© 2019 UPV/EHU Attribution-NonCommercial-ShareAlike 4.0 International
dc.identifier.doi10.1387/ekaia.19724


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

© 2019 UPV/EHU Attribution-NonCommercial-ShareAlike 4.0 International
Except where otherwise noted, this item's license is described as © 2019 UPV/EHU Attribution-NonCommercial-ShareAlike 4.0 International