Nexus#: a distributed hardware task manager for task-based programming models
dc.contributor.author | Dallou, Tamer | |
dc.contributor.author | Engelhardt, Nina | |
dc.contributor.author | Elhossini, Ahmed | |
dc.contributor.author | Juurlink, Ben | |
dc.date.accessioned | 2017-03-09T08:30:26Z | |
dc.date.available | 2017-03-09T08:30:26Z | |
dc.date.issued | 2015 | |
dc.description.abstract | In the era of multicore systems, it is expected that the number of cores that can be integrated on a single chip will be 3-digit. The key to utilize such a huge computational power is to extract the very fine parallelism in the user program. This is non-trivial for the average programmer, and becomes very hard as the number of potential parallel instances increases. Task-based programming models such as OmpSs are promising, since they handle the detection of dependencies and synchronization for the programmer. However, state-of-the-art research shows that task management is not cheap, and introduces a significant overhead that limits the scalability of OmpSs. Nexus# is a hardware accelerator for the OmpSs runtime system, which dynamically monitors dependencies between tasks. It is fully synthesizable in VHDL, and has a distributed task graph model to achieve the best scalability. Supporting tasks with arbitrary number of parameters and any dependency pattern, Nexus# achieves better performance than Nanos, the official OmpSs runtime system, and scales well for the H264dec benchmark with very fine grained tasks, among other benchmarks from the Starbench suite. | en |
dc.identifier.isbn | 978-1-4799-8650-7 | |
dc.identifier.issn | 1530-2075 | |
dc.identifier.uri | https://depositonce.tu-berlin.de/handle/11303/6228 | |
dc.identifier.uri | http://dx.doi.org/10.14279/depositonce-5787 | |
dc.language.iso | en | |
dc.rights.uri | http://rightsstatements.org/vocab/InC/1.0/ | |
dc.subject.ddc | 004 Datenverarbeitung; Informatik | |
dc.subject.other | data flow | en |
dc.subject.other | hardware support | en |
dc.subject.other | hardware task scheduler | en |
dc.subject.other | parallel programming | en |
dc.subject.other | task graph | en |
dc.subject.other | task manager | en |
dc.title | Nexus#: a distributed hardware task manager for task-based programming models | en |
dc.type | Conference Object | en |
dc.type.version | acceptedVersion | en |
dcterms.bibliographicCitation.doi | 10.1109/IPDPS.2015.79 | |
dcterms.bibliographicCitation.originalpublishername | IEEE | en |
dcterms.bibliographicCitation.originalpublisherplace | New York, NY [u.a.] | en |
dcterms.bibliographicCitation.pageend | 1138 | |
dcterms.bibliographicCitation.pagestart | 1129 | |
dcterms.bibliographicCitation.proceedingstitle | 2015 IEEE International Parallel and Distributed Processing Symposium : IPDPS | en |
tub.accessrights.dnb | domain | |
tub.affiliation | Fak. 4 Elektrotechnik und Informatik::Inst. Technische Informatik und Mikroelektronik::FG Architektur eingebetteter Systeme | de |
tub.affiliation.faculty | Fak. 4 Elektrotechnik und Informatik | de |
tub.affiliation.group | FG Architektur eingebetteter Systeme | de |
tub.affiliation.institute | Inst. Technische Informatik und Mikroelektronik | de |
tub.publisher.universityorinstitution | Technische Universität Berlin | en |
Files
Original bundle
1 - 1 of 1