Efficient data redistribution for malleable applications
![Thumbnail](/xmlui/bitstream/handle/10234/205853/88622.pdf.jpg?sequence=4&isAllowed=y)
Visualitza/
Impacte
![Google Scholar](/xmlui/themes/Mirage2/images/uji/logo_google.png)
![Microsoft Academico](/xmlui/themes/Mirage2/images/uji/logo_microsoft.png)
Metadades
Mostra el registre complet de l'elementcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7036
comunitat-uji-handle3:10234/146069
comunitat-uji-handle4:
INVESTIGACIONMetadades
Títol
Efficient data redistribution for malleable applicationsData de publicació
2023Editor
ACM Digital Library; Association for Computing MachineryISBN
979840070785Cita bibliogràfica
MARTÍN ÁLVAREZ, Iker, et al. Efficient data redistribution for malleable applications. In: Proceedings of the SC'23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis. 2023. p. 416-426Tipus de document
info:eu-repo/semantics/conferenceObjectVersió de l'editorial
https://dl.acm.org/doi/10.1145/3624062.3624110Versió
info:eu-repo/semantics/publishedVersionParaules clau / Matèries
Resum
Process malleability can be defined as the ability of a distributed MPI parallel job to change the number of processes on–the–fly without stopping its execution, reallocating the compute resources originally assigned ... [+]
Process malleability can be defined as the ability of a distributed MPI parallel job to change the number of processes on–the–fly without stopping its execution, reallocating the compute resources originally assigned to the job, and without storing application data to disk. MPI malleability consists of four stages: resource reallocation, process management, data redistribution and execution resuming. Among them, data redistribution is the most time-consuming and determines the reconfiguration time. In this paper, we compare different implementations of this stage using point-to-point and collective MPI operations, and discuss the impact of overlapping computation-communication. We then combine these strategies with different methods to expand/shrink jobs, using a synthetic application to emulate MPI-based codes and their malleable counterparts, in order to evaluate the effect of different malleability methods in parallel distributed applications. The results show that the use of asynchronous techniques speeds up execution by 1.14 and 1.21, depending on the network used. [-]
Descripció
In Workshops of The International Conference on High Performance Computing, Network, Storage,
and Analysis (SC-W 2023), November 12–17, 2023, Denver, CO, USA. ACM, New York, NY, USA.
Publicat a
In: Proceedings of the SC'23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis. 2023. p. 416-426.Entitat finançadora
Ministerio de Ciencia, Innovación y Universidades | Agencia Estatal de Investigación | Generalitat Valenciana | Comisión Europea | Unión Europea | Ministerio de Pesca | European Union NextGenera-tionEU/PRTR
Codi del projecte o subvenció
PID2020-113656RB-C21 | ACIF/2021/260 | 955606 | PCI2021-121958
Drets d'accés
info:eu-repo/semantics/openAccess