Distributed-Memory load balancing with cyclic token-based work-stealing applied to reverse time migration

Martins, Idalmis Milian Sardina; Assis, Ítalo Augusto Souza; Oliveira, Antônio D. S.; Barros, Tiago; Bianchini, Calebe P.; Souza, Samuel Xavier de

Distributed-Memory load balancing with cyclic token-based work-stealing applied to reverse time migration

Página do item simplificado Estatísticas

dc.contributor.author	Martins, Idalmis Milian Sardina
dc.contributor.author	Assis, Ítalo Augusto Souza
dc.contributor.author	Oliveira, Antônio D. S.
dc.contributor.author	Barros, Tiago
dc.contributor.author	Bianchini, Calebe P.
dc.contributor.author	Souza, Samuel Xavier de
dc.date.accessioned	2020-12-11T15:21:35Z
dc.date.available	2020-12-11T15:21:35Z
dc.date.issued	2019
dc.description.resumo	Reverse time migration (RTM) is a prominent technique in seismic imaging. Its resulting subsurface images are used in the industry to investigate with higher confidence the existence and the conditions of oil and gas reservoirs. Because of its high computational cost, RTM must make use of parallel computers. Balancing the workload distribution of an RTM is a growing challenge in distributed computing systems. The competition for shared resources and the differently-sized tasks of the RTM are some of the possible sources of load imbalance. Although many load balancing techniques exist, scaling up for large problems and large systems remains a challenge because synchronization overhead also scales. This paper proposes a cyclic token-based work-stealing (CTWS) algorithm for distributed memory systems applied to RTM. The novel cyclic token approach reduces the number of failed steals, avoids communication overhead, and simplifies the victim selection and the termination strategy. The proposed method is implemented as a C library using the one-sided communication feature of the message passing interface (MPI) standard. Results obtained by applying the proposed technique to balance the workload of a 3D RTM system present a factor of 14.1 % speedup and reductions of the load imbalance of 78.4 % when compared to the conventional static distribution	pt_BR
dc.identifier.citation	ASSIS, Italo A. S.; OLIVEIRA, Antonio D. S.; BARROS, Tiago; SARDINA, Idalmis M.; BIANCHINI, Calebe P.; DE-SOUZA, Samuel Xavier. Distributed-Memory Load Balancing With Cyclic Token-Based Work-Stealing Applied to Reverse Time Migration. Ieee Access, [S.L.], v. 7, p. 128419-128430, 2019. Disponível em: https://ieeexplore.ieee.org/document/8822671/. Acesso em: 15 out. 2020. http://dx.doi.org/10.1109/access.2019.2939100.	pt_BR
dc.identifier.doi	10.1109/ACCESS.2019.2939100
dc.identifier.issn	2169-3536
dc.identifier.uri	https://repositorio.ufrn.br/handle/123456789/30958
dc.language	en	pt_BR
dc.publisher	Institute of Electrical and Electronics Engineers	pt_BR
dc.rights	Attribution 3.0 Brazil	*
dc.rights.uri	http://creativecommons.org/licenses/by/3.0/br/	*
dc.subject	Load balancing	pt_BR
dc.subject	Reverse time migration	pt_BR
dc.subject	Work-stealing	pt_BR
dc.subject	One-sided communication	pt_BR
dc.subject	Distributed memory	pt_BR
dc.title	Distributed-Memory load balancing with cyclic token-based work-stealing applied to reverse time migration	pt_BR
dc.type	article	pt_BR