Easy use of high performance computers for fusion simulations

Frauel, Y.; Coster, D.; Guillerminet, B.; Imbeaux, F.; Jackson, Adrian; Konz, C.; Owsiak, Michał; Płóciennik, Marcin; Scott, B.; Strand, Pär

doi:10.1016/j.fusengdes.2012.04.015

Cited by 6 publications

(8 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The multiscale models consist of the following:

Tied multiscale models. A tokamak plasma model (Transport Turbulence Equilibrium, TTE) from the fusion community [21] and a cerebrovascular blood flow model (HemeLB) from the biomedical community [23].

Scalable multiscale models.

…”

Section: Resultsmentioning

confidence: 99%

“…This is motivated by the recently completed MAPPER project, 1 which aimed to facilitate large multiscale simulations on distributed e-Infrastructure. The project was driven by seven multiscale applications from the following disciplines: nano materials [ 22 ], fusion [ 21 ], biomedicine [ 23 ], hydrology [ 29 ] and systems biology [ 20 ]. We divide these applications into three categories based on how they may benefit from distributed computing: (i) by increasing simulation speed by supplementing local dependencies (e.g.…”

Section: Introductionmentioning

confidence: 99%

“…Over the past few years, we have developed a large collection of multiscale models [19][20][21][22][23][24] and have found that these multiscale models are computationally intensive. Other examples of such demanding multiscale models are Earth system models [9,[25][26][27], each taking a componentbased approach with the possibility for distributed computing.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Performance of distributed multiscale simulations

Borgdorff

Belgacem

Bona-Casas

et al. 2014

Phil. Trans. R. Soc. A.

Self Cite

View full text Add to dashboard Cite

Multiscale simulations model phenomena across natural scales using monolithic or component-based code, running on local or distributed resources. In this work, we investigate the performance of distributed multiscale computing of component-based models, guided by six multiscale applications with different characteristics and from several disciplines. Three modes of distributed multiscale computing are identified: supplementing local dependencies with large-scale resources, load distribution over multiple resources, and load balancing of small- and large-scale resources. We find that the first mode has the apparent benefit of increasing simulation speed, and the second mode can increase simulation speed if local resources are limited. Depending on resource reservation and model coupling topology, the third mode may result in a reduction of resource consumption.

show abstract

“…The multiscale models consist of the following:

Scalable multiscale models.

…”

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Performance of distributed multiscale simulations

Borgdorff

Belgacem

Bona-Casas

et al. 2014

Phil. Trans. R. Soc. A.

Self Cite

View full text Add to dashboard Cite

show abstract

“…GPU) or use MPI and/or OpenMP for their parallelization. Solutions have been developed in the EU-ITM project to enable the usage of MPI and OpenMP for Kepler workflow components, together with various distributed computing strategies (remote submission on grid or high performance computers, web services) [17,18]. These solutions could be ported in the near future to the IMAS infrastructure to optimize the performance of various applications.…”

Section: Fine-grained Component Integrationmentioning

confidence: 99%

Design and first applications of the ITER integrated modelling & analysis suite

Imbeaux¹,

Pinches²,

Lister³

et al. 2015

Nucl. Fusion

Self Cite

121

View full text Add to dashboard Cite

The ITER Integrated Modelling & Analysis Suite (IMAS) will support both plasma operation and research activities on the ITER tokamak experiment. The IMAS will be accessible to all ITER members as a key tool for the scientific exploitation of ITER. The backbone of the IMAS infrastructure is a standardized, machine-generic data model that represents simulated and experimental data with identical structures. The other outcomes of the IMAS design and prototyping phase are a set of tools to access data and design integrated modelling workflows, as well as first plasma simulators workflows and components implemented with various degrees of modularity.

show abstract

“…So far MUSCLE 2 is being used in a number of multiscale models, for instance a collection of parallel Fortran codes of the Fusion community [18], a gene regulatory network simulation [38], a hydrology application [5], and in a multiscale model of in-stent restenosis [10,6,20].…”

Section: Introductionmentioning

confidence: 99%

Distributed multiscale computing with MUSCLE 2, the Multiscale Coupling Library and Environment

Borgdorff

Mamonski

Bosak

et al. 2014

Journal of Computational Science

View full text Add to dashboard Cite

We present the Multiscale Coupling Library and Environment: MUSCLE 2. This multiscale component-based execution environment has a simple to use Java, C++, C, Python and Fortran API, compatible with MPI, OpenMP and threading codes. We demonstrate its local and distributed computing capabilities and compare its performance to MUSCLE 1, file copy, MPI, MPWide, and GridFTP. The local throughput of MPI is about two times higher, so very tightly coupled code should use MPI as a single submodel of MUSCLE 2; the distributed performance of GridFTP is lower, especially for small messages. We test the performance of a canal system model with MUSCLE 2, where it introduces an overhead as small as 5% compared to MP

show abstract

Easy use of high performance computers for fusion simulations

Cited by 6 publications

References 8 publications

Performance of distributed multiscale simulations

Performance of distributed multiscale simulations

Design and first applications of the ITER integrated modelling & analysis suite

Distributed multiscale computing with MUSCLE 2, the Multiscale Coupling Library and Environment

Contact Info

Product

Resources

About