Large scale in transit computation of quantiles for ensemble runs

Abstract : The classical approach for quantiles computation requires availability of the full sample before ranking it. In uncertainty quantification of numerical simulation models, this approach is not suitable at exascale as large ensembles of simulation runs would need to gather a prohibitively large amount of data. This problem is solved thanks to an on-the-fly and iterative approach based on the Robbins-Monro algorithm. This approach relies on Melissa, a file avoiding, adaptive, fault-tolerant and elastic framework. On a validation case producing 11 TB of data, which consists in 3000 fluid dynamics parallel simulations on a 6M cell mesh, it allows on-line computation of spatio-temporal maps of percentiles.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [20 references]  Display  Hide  Download

https://hal.inria.fr/hal-02016828
Contributor : Bertrand Iooss <>
Submitted on : Friday, May 10, 2019 - 3:20:30 PM
Last modification on : Sunday, May 12, 2019 - 1:03:07 AM

Files

paper.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02016828, version 2
  • ARXIV : 1905.04180

Citation

Alejandro Ribes, Théophile Terraz, Bertrand Iooss, Yvan Fournier, Bruno Raffin. Large scale in transit computation of quantiles for ensemble runs. 2019. ⟨hal-02016828v2⟩

Share

Metrics

Record views

80

Files downloads

178