Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 37
Encontrada(s) 1 página(s)

DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Heat loss prediction of a confined premixed jet flame using a conjugate heat transfer approach

  • Gövert, Simon
  • Mira, Daniel
  • Zavala Ake, Miguel
  • Kok, Jim B.W.
  • Vázquez, Mariano
  • Houzeaux, Guillaume
The presented work addresses the investigation of the heat loss of a confined turbulent jet flame in a lab-scale combustor using a conjugate-heat transfer approach and large-eddy simulation. The analysis includes the assessment of the principal mechanisms of heat transfer in this combustion chamber: radiation, convection and conduction of heat over walls. A staggered approach is used to couple the reactive flow field to the heat conduction through the solid and both domains are solved using two implementations of the same code. Numerical results are compared against experimental data and an assessment of thermal boundary conditions to improve the prediction of the reactive flow field is given. © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CCBY license (http://creativecommons.org/licenses/by/4.0/)., The research leading to these results has received funding through the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (FP7, 2007–2013) under the Grant agreement No. FP7-290042 for the project COPA-GT as well as the European Union’s Horizon 2020 Programme (2014–2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project, Grant agreement No. 689772., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

The Effect of Partial Premixing and Heat Loss on the Reacting Flow Field Prediction of a Swirl Stabilized Gas Turbine Model Combustor

  • Gövert, Simon
  • Mira, Daniel
  • Kok, Jim B.W.
  • Vázquez, Mariano
  • Houzeaux, Guillaume
This work addresses the prediction of the reacting flow field in a swirl stabilized gas turbine model combustor using large-eddy simulation. The modeling of the combustion chemistry is based on laminar premixed flamelets and the effect of turbulence-chemistry interaction is considered by a presumed shape probability density function. The prediction capabilities of the presented combustion model for perfectly premixed and partially premixed conditions are demonstrated. The effect of partial premixing for the prediction of the reacting flow field is assessed by comparison of a perfectly premixed and partially premixed simulation. Even though significant mixture fraction fluctuations are observed, only small impact of the non-perfect premixing is found on the flow field and flame dynamics. Subsequently, the effect of heat loss to the walls is assessed assuming perfectly premixing. The adiabatic baseline case is compared to heat loss simulations with adiabatic and non-adiabatic chemistry tabulation. The results highlight the importance of considering the effect of heat loss on the chemical kinetics for an accurate prediction of the flow features. Both heat loss simulations significantly improve the temperature prediction, but the non-adiabatic chemistry tabulation is required to accurately capture the chemical composition in the reacting layers. © 2017, The Author(s)., Funding The research leading to these results has received funding through the People Program (Marie Curie Actions) of the European Union’s Seventh Framework Program (FP7, 2007-2013) for the project COPA-GT (grant agreement No. FP7-290042) as well as the European Union’s Horizon 2020 Program (2014-2020) and the Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (grant agreement No. 689772). Furthermore, computer resources and technical assistance has been provided by the Red Española de Supercomputación (RES) (grant numbers FI-2015-1-0002, FI-2015-3-0013, FI-2016-1-0001)., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Quality Control of Surface Wind Observations in Northeastern North America. Part II: Measurement Errors

  • Lucio-Eceiza, Etor E.
  • González-Rouco, J. F.
  • Navarro, Jorge
  • Beltrami, H.
  • Conte, Jorge
A quality control (QC) process has been developed and applied to an observational database of surface wind speed and wind direction in northeastern North America. The database combines data from three datasets of different initial quality, including a total of 526 land stations and buoys distributed over the provinces of eastern Canada and five adjacent northeastern U.S. states. The data span from 1953 to 2010. The first part of the QC deals with data management issues and is developed in a companion paper. Part II, presented herein, is focused on the detection of measurement errors and deals with low-variability errors, like the occurrence of unrealistically long calms, and high-variability problems, like rapid changes in wind speed; some types of biases in wind speed and wind direction are also considered. About 0.5% (0.16%) of wind speed (wind direction) records have been flagged. Additionally, 15.87% (1.73%) of wind speed (wind direction) data have been corrected. The most pervasive error type in terms of affected sites and erased data corresponds to unrealistic low wind speeds (89% of sites affected with 0.35% records removed). The amount of detected and corrected/removed records in Part II (~9%) is approximately two orders of magnitude higher than that of Part I. Both management and measurement errors are shown to have a discernible impact on the statistics of the database., EELE was supported by the Agreement of Cooperation 4164281 between the UCM and St. Francis Xavier University, and projects CGL2014- 59644-R and PCIN-2014-017-C07-06 of the MINECO (Spain). Funding for 4164281 was provided by the Natural Sciences and Engineering Research Council of Canada (NSERC DG 140576948), the Canada Research Chairs Program (CRC 230687), and the Atlantic Innovation Fund (AIF-ACOA). HB holds a Canada Research Chair in Climate Dynamics. JN and JFGR were supported by projects PCIN-2014-017-C07-03, PCIN-2014-017-C07-06, CGL2011-29677-C02-01, and CGL2011-29677-C02-02 of the MINECO (Spain). JC was supported by Global Forecasters until March 2014. This research has been conducted under the Joint Research Unit between UCM and CIEMAT, by the Collaboration Agreement 7158/ 2016. The research has also received funding from the European Union’s Horizon 2020 Programme (2014–2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (www.hpc4e.eu), Grant Agreement 689772., Peer reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

A CFD framework for offshore and onshore wind farm simulation

  • Avila, Matias
  • Gargallo-Peiró, Abel
  • Folch, Arnau|||0000-0002-0677-6366
We present a wind simulation framework for offshore and onshore wind farms. The simulation framework involves an automatic hybrid high-quality mesh generation process, a pre-processing to impose initial and boundary conditions, and a solver for the Reynolds Averaged Navier-Stokes (RANS) equations with two different turbulence models, a modified standard k-epsilon model and a realizable k-epsilon model in which we included the Coriolis effects. Wind turbines are modeled as actuator discs. The wind farm simulation framework has been implemented in Alya, an in-house High Performance Computing (HPC) multi-physics finite element parallel solver. An application example is shown for an onshore wind farm composed of 165 turbines., This work has been supported by the EU H2020 projects New European Wind Atlas ERA-NET PLUS (NEWA), High Performance Computing for Energy (HPC4E, grant agreement 689772), and the Energy oriented Centre of Excellence (EoCoE, grant agreement 676629). We also thank two anonymous reviewers for their constructive comments on the first version of the manuscript., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Mesh generation, sizing and convergence for onshore and offshore wind farm Atmospheric Boundary Layer flow simulation with actuator discs

  • Gargallo Peiró, Abel|||0000-0003-3742-2197
  • Avila, Matias
  • Owen, Herbert
  • Prieto-Godino, Luis
  • Folch, Arnau|||0000-0002-0677-6366
A new mesh generation process for wind farm modeling is presented together with a mesh convergence and sizing analysis for wind farm flow simulations. The generated meshes are tailored to simulate Atmospheric Boundary Layer (ABL) flows on complex terrains modeling the wind turbines as actuator discs. The wind farm mesher is fully automatic and, given the topography and the turbine characteristics (location, diameter and hub height), it generates a hybrid mesh conformal with the actuator discs and refined upwind and downstream. Moreover, it presents smooth element size transitions across scales and avoids extending high-resolution areas to all the domain. We take advantage of our automatic and robust mesher to study the mesh convergence of our RANS solver with linear elements, obtaining quadratic mesh convergence for a quantity of interest in all the tested cases. In addition, we quantify the mesh resolution at the terrain surface and at the actuator discs required to achieve a given numerical error in simulations in onshore and offshore frameworks. Finally, we present the generated meshes and the simulation results for an offshore and an onshore wind farm. We analyze in detail one particular wind direction for both cases, and for the onshore wind farm we use our automatic framework to estimate the yearly production of energy and measuring the error against the actual produced one., This work has been partially funded by the EU H2020Energy oriented Center of Excellence (EoCoE) for computer applications, the New European Wind Atlas (NEWA) and the High Performance Computing for Energy (HPC4E) projects. We thank Iberdrola Renovables for their collaboration and for providing wind farm data to validate the developed techniques., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Acceleration strategies for elastic full waveform inversion workflows in 2D and 3D

  • Kormann, Jean
  • Rodríguez, Juan E.
  • Ferrer, Miguel
  • Farrés, Albert
  • Gutiérrez, Natalia|||0000-0001-7921-7322
  • de la Puente, Josep
  • Hanzich, Mauricio
  • Cela, José M.
Full waveform inversion (FWI) is one of the most challenging procedures to obtain quantitative information of the subsurface. For elastic inversions, when both compressional and shear velocities have to be inverted, the algorithmic issue becomes also a computational challenge due to the high cost related to modelling elastic rather than acoustic waves. This shortcoming has been moderately mitigated by using high-performance computing to accelerate 3D elastic FWI kernels. Nevertheless, there is room in the FWI workflows for obtaining large speedups at the cost of proper grid pre-processing and data decimation techniques. In the present work, we show how by making full use of frequency-adapted grids, composite shot lists and a novel dynamic offset control strategy, we can reduce by several orders of magnitude the compute time while improving the convergence of the method in the studied cases, regardless of the forward and adjoint compute kernels used., The authors thank REPSOL for the permission to publish the present research and for funding through the AURORA project. J. Kormann also thankfully acknowledges the computer resources, technical expertise and assistance provided by the Barcelona Supercomputing Center - Centro Nacional de Supercomputacti ´on together with the Spanish Supercomputing Network (RES) through grant FI-2014-2-0009. This project has received funding from the European Union’s Horizon 2020, research and innovation programme under the Marie Skłodowska-Curie grant agreement no. 644202. The research leading to these results has received funding from the European Union’s Horizon 2020 Programme (2014–2020) and from the Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (www.hpc4e.eu), grant agreement no. 689772.We further want to thank the Editor Clint N. Dawson for his help, and Andreas Fichtner and an anonymous reviewer for their comments and suggestions to improve the manuscript., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

A wind field downscaling strategy based on domain segmentation and transfer functions

  • Barcons, Jordi
  • Avila, Matias
  • Folch, Arnau|||0000-0002-0677-6366
This paper presents a novel methodology for mesoscale‐to‐microscale downscaling of near‐surface wind fields. The model chain consists on the Weather Research and Forecast mesoscale model and the Alya‐CFDWind microscale model (assuming neutral stability). The downscaling methodology combines precomputed microscale simulations with a mesoscale forecast using a domain segmentation technique and transfer functions. As a result, the downscaled wind field preserves the mesoscale pattern but, at the same time, incorporates local mesoscale subgrid terrain effects, particularly at valleys and channelling zones. The methodology has been validated over a 9‐month period on a very complex terrain site instrumented with a dense observational network of meteorological masts. With respect to mesoscale results, the global skills of the downscaled wind at masts improve for wind direction and remain similar for wind velocity. However, a substantial improvement occurs under stable and neutral conditions and for high wind velocity regimes., This work has been partially funded by the High Performance Computing for Energy (HPC4E) project (call H2020-EUB-2015, Topic: EUB-2-2015, type of action RIA, Grant Agreement number 689772) and the SEDAR ("Simulación eólica de alta resolución") project. It has also been partially supported by the Energy-oriented Centre of Excellence (EoCoE) (Grant Agreement number 676629, funded within the H2020 framework of the EuropeanUnion). J.B. is grateful to a PhD fellowship from the IndustrialDoctorates Plan of the Government of Catalonia (Ref. eco/2497/2013). We also thank Daniel Paredes and Luis Prieto from Iberdrola Renovables S.A. for providing us access to met masts data for validation., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Optimization strategies for geophysics models on manycore systems

  • Serpa, Matheus S.
  • Cruz, Eduardo HM
  • Diener, Matthias
  • Krause, Arthur M.
  • Navaux, Philippe
  • Panetta, Jairo
  • Farrés Coma, Albert
  • Rosas, Claudia|||0000-0002-7814-0359
  • Hanzich, Mauricio
Many software mechanisms for geophysics exploration in oil and gas industries are based on wave propagation simulation. To perform such simulations, state-of-the-art high-performance computing architectures are employed, generating results faster with more accuracy at each generation. The software must evolve to support the new features of each design to keep performance scaling. Furthermore, it is important to understand the impact of each change applied to the software to improve the performance as most as possible. In this article, we propose several optimization strategies for a wave propagation model for six architectures: Intel Broadwell, Intel Haswell, Intel Knights Landing, Intel Knights Corner, NVIDIA Pascal, and NVIDIA Kepler. We focus on improving the cache memory usage, vectorization, load balancing, portability, and locality in the memory hierarchy. We analyze the hardware impact of the optimizations, providing insights of how each strategy can improve the performance. The results show that NVIDIA Pascal outperforms the other considered architectures by up to 8.5×., Our research received funding from the EU H2020 Programme and from MCTI/RNP-Brazil under the HPC4E project, grant agreement 689772, as well as from CNPq and Capes., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Diurnal cycle RANS simulations applied to wind resource assessment

  • Barcons, Jordi
  • Avila, Matias
  • Folch, Arnau|||0000-0002-0677-6366
Microscale computational fluid dynamics (CFD) models can be used for wind resource assessment on complex terrains. These models generally assume neutral atmospheric stratification, an assumption that can lead to inaccurate modeling results and to large uncertainties at certain sites. We propose a methodology for wind resource evaluation based on unsteady Reynolds averaged Navier‐Stokes (URANS) simulations of diurnal cycles including the effect of thermal stratification. Time‐dependent boundary conditions are generated by a 1D precursor to drive 3D diurnal cycle simulations for a given geostrophic wind direction sector. Time instants of the cycle representative of four thermal stability regimes are sampled within diurnal cycle simulations and combined with masts time series to obtain the wind power density (WPD). The methodology has been validated on a complex site instrumented with seven met masts. The WPD spatial distribution is in good agreement with observations with the mean absolute error improving 17.1% with respect to the neutral stratification assumption., This work has been partially supported by the three EU H2020 projects, New European Wind Atlas ERA‐NET PLUS (NEWA, FP7‐ENERGY.2013.10.1.2, European Commission's grant agreement 618122), High Performance Computing for Energy (HPC4E, grant agreement 689772), and the Energy oriented Centre of Excellence (EoCoE, grant agreement 676629), and the SEDAR (“Simulación eólica de alta resolución”) project. Jordi Barcons is grateful to a PhD fellowship from the Industrial Doctorates Plan of the Government of Catalonia (Ref. eco/2497/2013). We also thank Iberdrola Renovables Energa S.A. and Impulsora Latinoamericana de Energa Renovables S.A. for providing the access to Puebla met masts data for validation and to Luis Prieto and Daniel Paredes for their help. We also thank the reviewers for their productive comments and observations., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

The Effect of Partial Premixing and Heat Loss on the Reacting Flow Field Prediction of a Swirl Stabilized Gas Turbine Model Combustor

  • Gövert, Simon
  • Mira Martínez, Daniel
  • Kok, Jim B. W.
  • Vázquez, Mariano|||0000-0002-2526-6708
  • Houzeaux, Guillaume|||0000-0002-2592-1426
This work addresses the prediction of the reacting flow field in a swirl stabilized gas turbine model combustor using large-eddy simulation. The modeling of the combustion chemistry is based on laminar premixed flamelets and the effect of turbulence-chemistry interaction is considered by a presumed shape probability density function. The prediction capabilities of the presented combustion model for perfectly premixed and partially premixed conditions are demonstrated. The effect of partial premixing for the prediction of the reacting flow field is assessed by comparison of a perfectly premixed and partially premixed simulation. Even though significant mixture fraction fluctuations are observed, only small impact of the non-perfect premixing is found on the flow field and flame dynamics. Subsequently, the effect of heat loss to the walls is assessed assuming perfectly premixing. The adiabatic baseline case is compared to heat loss simulations with adiabatic and non-adiabatic chemistry tabulation. The results highlight the importance of considering the effect of heat loss on the chemical kinetics for an accurate prediction of the flow features. Both heat loss simulations significantly improve the temperature prediction, but the non-adiabatic chemistry tabulation is required to accurately capture the chemical composition in the reacting layers., The research leading to these results has received funding through the People Program (Marie Curie Actions) of the European Union’s Seventh Framework Program (FP7, 2007-2013) for the project COPA-GT (grant agreement No. FP7-290042) as well as the European Union’s Horizon 2020 Program (2014-2020) and the Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (grant agreement No. 689772). Furthermore, computer resources and technical assistance has been provided by the Red Española de Supercomputación (RES) (grant numbers FI-2015-1-0002, FI-2015-3-0013, FI-2016-1-0001)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Toward an automatic full-wave inversion: Synthetic study cases

  • Kormann, Jean
  • Rodriguez, Juan E.
  • Gutiérrez, Natalia|||0000-0001-7921-7322
  • Ferrer, Miguel
  • Rojas, Otilio
  • de la Puente, Josep
  • Hanzich, Mauricio
  • Cela, Jose M.
Full-waveform inversion (FWI) in seismic scenarios continues to be a complex procedure for subsurface imaging that might require extensive human interaction in terms of model setup, constraints, and data preconditioning. The underlying reason is the strong nonlinearity of the problem that forces the addition of a priori knowledge (or bias) in order to obtain geologically sound results. In particular, when the use of a long-offset receiver is not possible or may not favor the reconstruction of the fine structure of the model, one needs to rely on reflection data. As a consequence, the inversion process is more prone to becoming stuck in local minima. Nevertheless, misfit functionals can be devised that can either cope with missing long-wavenumber features of initial models (e.g., cross-correlation-based misfit) or invert reflection-dominated data whenever the models are sufficiently good (e.g., normalized offset-limited least-squares misfit). By combining both, high-frequency data content with poor initial models can be successfully inverted. If one can figure out simple parameterizations for such functionals, the amount of uncertainty and manual work related to tuning FWI would be substantially reduced. Thus, FWI might become a semiautomatized imaging tool., We want to thank Repsol for funding this research by means of the Aurora project. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 644202. Additionally, the research leading to these results has received funding from the European Union’s Horizon 2020 Programme (2014-2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (www.hpc4e.eu), grant agreement No 689772. We acknowledge Chevron for the dataset that was used in our second example., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

MPI+X: task-based parallelization and dynamic load balance of finite element assembly

  • Garcia-Gasulla, Marta
  • Houzeaux, Guillaume|||0000-0002-2592-1426
  • Ferrer, Roger
  • Artigues, Antoni
  • López, Victor
  • Labarta Mancho, Jesús José|||0000-0002-7489-4727
  • Vázquez, Mariano|||0000-0002-2526-6708
The main computing tasks of a finite element code(FE) for solving partial differential equations (PDE's) are the algebraic system assembly and the iterative solver. This work focuses on the first task, in the context of a hybrid MPI+X paradigm. Although we will describe algorithms in the FE context, a similar strategy can be straightforwardly applied to other discretization methods, like the finite volume method. The matrix assembly consists of a loop over the elements of the MPI partition to compute element matrices and right-hand sides and their assemblies in the local system to each MPI partition. In a MPI+X hybrid parallelism context, X has consisted traditionally of loop parallelism using OpenMP. Several strate- gies have been proposed in the literature to implement this loop parallelism, like coloring or substructuring techniques to circumvent the race condition that appears when assembling the element system into the local system. The main drawback of the first technique is the decrease of the IPC due to bad spatial locality. The second technique avoids this issue but requires extensive changes in the implementation, which can be cumbersome when several element loops should be treated. We propose an alternative, based on the task parallelism of the element loop using some extensions to the OpenMP programming model. The task- ification of the assembly solves both aforementioned problems. In addition, dynamic load balance will be applied using the DLB library, especially efficient in the presence of hybrid meshes, where the relative costs of the different elements is impossible to estimate a priori. This paper presents the proposed methodology, its implementation and its validation through the solution of large computational mechanics problems up to 16k cores., The use of large part of a supercomputer, even more in normal conditions of use, is never an innocuous exercise. The research leading to these results has received funding from: the European Union's Horizon 2020 Programme (2014–2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP), HPC4E Project, grant agreement 689772; the Energy oriented Centre of Excellence (EoCoE), grant agreement number 676629, funded within the Horizon2020 framework of the European Union; The Spanish Government (grant SEV2015-0493 of the Severo Ochoa Program); the Spanish Ministry of Science and Innovation (contract TIN2015-65316-P); the Generalitat de Catalunya (contract 2014-SGR-1051); the Intel-BSC Exascale Lab collaboration project. Comissió Interdepartamental de Recerca i Innovació Tecnológica(Interdepartmental Commission for Research and Technological Innovation), Sí
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Multi-objective scheduling of Scientific Workflows in multisite clouds

  • Liu, Ji
  • Pacitti, Esther
  • Valduriez, Patrick
  • Oliveira, Daniel de
  • Mattoso, Marta
Clouds appear as appropriate infrastructures for executing Scientific Workflows (SWfs). A cloud is typically made of several sites (or data centers), each with its own resources and data. Thus, it becomes important to be able to execute some SWfs at more than one cloud site because of the geographical distribution of data or available resources among different cloud sites. Therefore, a major problem is how to execute a SWf in a multisite cloud, while reducing execution time and monetary costs. In this paper, we propose a general solution based on multi-objective scheduling in order to execute SWfs in a multisite cloud. The solution consists of a multi-objective cost model including execution time and monetary costs, a Single Site Virtual Machine (VM) Provisioning approach (SSVP) and ActGreedy, a multisite scheduling approach. We present an experimental evaluation, based on the execution of the SciEvol SWf in Microsoft Azure cloud. The results reveal that our scheduling approach significantly outperforms two adapted baseline algorithms (which we propose by adapting two existing algorithms) and the scheduling time is reasonable compared with genetic and brute-force algorithms. The results also show that our cost model is accurate and that SSVP can generate better VM provisioning plans compared with an existing approach., Work partially funded by EU H2020 Programme and MCTI/RNP-Brazil (HPC4E grant agreement number 689772), CNPq, FAPERJ, and INRIA (MUSIC project), Microsoft (ZcloudFlow project) and performed in the context of the Computational Biology Institute (www.ibc-montpellier.fr). We would like to thank Kary Ocaña for her help in modeling and executing the SciEvol SWf., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Parallel mesh partitioning based on space filling curves

  • Borrell, Ricard
  • Cajas García, Juan Carlos
  • Mira Martínez, Daniel
  • Taha, A.
  • Koric, S.
  • Vázquez, Mariano|||0000-0002-2526-6708
  • Houzeaux, Guillaume|||0000-0002-2592-1426
Larger supercomputers allow the simulation of more complex phenomena with increased accuracy. Eventually this requires finer and thus also larger geometric discretizations. In this context, and extrapolating to the Exascale paradigm, meshing operations such as generation, deformation, adaptation/regeneration or partition/load balance, become a critical issue within the simulation workflow. In this paper we focus on mesh partitioning. In particular, we present a fast and scalable geometric partitioner based on Space Filling Curves (SFC), as an alternative to the standard graph partitioning approach. We have avoided any computing or memory bottleneck in the algorithm, while we have imposed that the solution achieved is independent (discounting rounding off errors) of the number of parallel processes used to compute it. The performance of the SFC-based partitioner presented has been demonstrated using up to 4096 CPU-cores in the Blue Waters supercomputer., The research leading to these results has received funding from the European Union Horizon 2020 Programme (2014-2020) and the Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (grant agreement No. 689772). This work is also part of the PRAC "Simulations of Aircraft Engines" supported by the National Science Foundation. It has also been financially supported by the PRACE preparatory access projects funded in part by the EU Horizon 2020 research and innovation programme (2014-2020) under grant agreement 653838. J.C. Cajas acknowledges the nancial sup- port of the `Consejo Nacional de Ciencia y Tecnolog a (CONACyT, M exico)' grant number 231588 290790. Ricard Borrell and Daniel Mira acknowledge the Juan de la Cierva postdoctoral grants with codes IJCI-2014-21034 and IJCI-2015-26686, respectively.
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

The combustion mitigation of methane as a non-CO2 greenhouse gas

  • Jian, X.
  • Mira Martínez, Daniel
  • Cluff, D.L.
Anthropogenic emissions of non-CO2 greenhouse gases such as fugitive methane contribute significantly to global warming. A review of fugitive methane combustion mitigation and utilisation technologies, which are primarily aimed at methane emissions from coal mining activities, with a focus on modelling and simulation of ultra-lean methane oxidation/combustion is presented. The challenges associated with ultra-lean methane oxidation are on the ignition of the ultra-lean mixture and sustainability of the combustion process. There is a lack of fundamental studies on chemical kinetics of ultra-lean methane combustion and reliable kinetic schemes that can be used together with computational fluid dynamics studies to design and develop advanced mitigation systems. Mitigation of methane as a greenhouse gas calls for more efforts on understanding ultra-lean combustion. Recuperative combustion provides a promising means for mitigating ultra-lean methane emissions. Progress is needed on effective methods to ignite and to recuperate and retain heat for oxidation/combustion of the ultra-lean mixtures. Catalysts can be very effective in reducing the temperatures required for oxidation while plasmas may be utilised to assist the ignition, but thermodynamic/aerodynamic limits of burning ultra-lean methane remain unexplored. Further technological developments may be focussed on developing innovative capturing technology as well as technological innovations to achieve effective ignition and sustainable oxidation/combustion., We would like to thank the reviewers and the editor for their constructive suggestions on the manuscript. These research results have received funding from the EU H2020 Programme (No. 689772) and from MCTI/RNP-Brazil under the HPC4E Project, grant agree- ment no 689772., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Hardware-Assisted Thread and Data Mapping in Hierarchical Multicore Architectures

  • Cruz, Eduardo H.M.
  • Diener, Matthias
  • Pilla, Laércio L.
  • Navaux, Philippe O.A.
The performance and energy efficiency of modern architectures depend on memory locality, which can be improved by thread and data mappings considering the memory access behavior of parallel applications. In this article, we propose intense pages mapping, a mechanism that analyzes the memory access behavior using information about the time the entry of each page resides in the translation lookaside buffer. It provides accurate information with a very low overhead. We present experimental results with simulation and real machines, with average performance improvements of 13.7% and energy savings of 4.4%, which come from reductions in cache misses and interconnection traffic., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

MPI+X: task-based parallelisation and dynamic load balance of finite element assembly

  • Garcia, Marta
  • Houzeaux, Guillaume|||0000-0002-2592-1426
  • Ferrer, Roger
  • Artigues, Antoni
  • López, Victor
  • Labarta Mancho, Jesús José|||0000-0002-7489-4727
  • Vázquez, Mariano|||0000-0002-2526-6708
The main computing phases of numerical methods for solving partial differential equations are the algebraic system assembly and the iterative solver. This work focuses on the first task, in the context of a hybrid MPI+X paradigm. The matrix assembly consists of a loop over the elements, faces, edges or nodes of the MPI partitions to compute element matrices and vectors and then of their assemblies. In a MPI+X hybrid parallelism context, X has consisted traditionally of loop parallelism using OpenMP, with different techniques to avoid the race condition, but presenting efficiency or implementation drawbacks. We propose an alternative, based on task parallelism using some extensions to the OpenMP programming model. In addition, dynamic load balance will be applied, especially efficient in the presence of hybrid meshes. This paper presents the proposed methodology, its implementation and its validation through the solution of large computational mechanics problems up to 16k cores., The research leading to these results has received funding from: the European Union's Horizon 2020 Programme (2014–2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP), HPC4E Project, grant agreement 689772; the Energy oriented Centre of Excellence (EoCoE), grant agreement number 676629, funded within the Horizon2020 framework of the European Union; The Spanish Government (grant SEV2015-0493 of the Severo Ochoa Program); the Spanish Ministry of Science and Innovation (contract TIN2015-65316-P); the Generalitat de Catalunya (contract 2014-SGR-1051); the Intel-BSC Exascale Lab collaboration project. Comissió Interdepartamental de Recerca i Innovació Tecnológica(Interdepartmental Commission for Research and Technological Innovation), Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Wavefield compression for adjoint methods in full-waveform inversion

  • Boehm, Christian
  • Hanzich, Mauricio
  • de la Puente, Josep
  • Fichtner, Andreas
Adjoint methods are a key ingredient of gradient-based full-waveform inversion schemes. While being conceptually elegant, they face the challenge of massive memory requirements caused by the opposite time directions of forward and adjoint simulations and the necessity to access both wavefields simultaneously for the computation of the sensitivity kernel. To overcome this bottleneck, we have developed lossy compression techniques that significantly reduce the memory requirements with only a small computational overhead. Our approach is tailored to adjoint methods and uses the fact that the computation of a sufficiently accurate sensitivity kernel does not require the fully resolved forward wavefield. The collection of methods comprises reinterpolation with a coarse temporal grid as well as adaptively chosen polynomial degree and floating-point precision to represent spatial snapshots of the forward wavefield on hierarchical grids. Furthermore, the first arrivals of adjoint waves are used to identify “shadow zones” that do not contribute to the sensitivity kernel. Numerical experiments show the high potential of this approach achieving an effective compression factor of three orders of magnitude with only a minor reduction in the rate of convergence. Moreover, it is computationally cheap and straightforward to integrate in finite-element wave propagation codes with possible extensions to finite-difference methods., The authors would like to thank Editor John Etgen, Associate Editor Anatoly Baumstein and three anonymous reviewers for their valuable comments and excellent suggestions to improve the paper. We gratefully acknowledge support by the Swiss National Supercomputing Centre (CSCS) in the form of CHRONOS project ch1 and PASC project GeoScale. The research leading to these results has received funding from the European Union's Horizon 2020 Programme (2014-2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (www.hpc4e.eu), grant agreement n. 689772. Furthermore, the first author acknowledges funding from Shell within the project \Boosting full-waveform inversion"., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Heat loss prediction of a confined premixed jet flame using a conjugate heat transfer approach

  • Gövert, S.
  • Mira Martínez, Daniel
  • Zavala-Ake, M.
  • Kok, J.B.W.
  • Vázquez, Mariano|||0000-0002-2526-6708
  • Houzeaux, Guillaume|||0000-0002-2592-1426
The presented work addresses the investigation of the heat loss of a confined turbulent jet flame in a lab-scale combustor using a conjugate-heat transfer approach and large-eddy simulation. The analysis includes the assessment of the principal mechanisms of heat transfer in this combustion chamber: radiation, convection and conduction of heat over walls. A staggered approach is used to couple the reactive flow field to the heat conduction through the solid and both domains are solved using two implementations of the same code. Numerical results are compared against experimental data and an assessment of thermal boundary conditions to improve the prediction of the reactive flow field is given., The research leading to these results has received funding through the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (FP7, 2007–2013) under the Grant agreement No. FP7-290042 for the project COPA-GT as well as the European Union’s Horizon 2020 Programme (2014–2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project, Grant agreement No. 689772. The authors thankfully acknowledge the computer resources, technical expertise and assistance provided by the Red Española de Supercomputación (RES). Finally, the authors would like to thank O. Lammel for the useful discussions and kindly providing the data for the comparison., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

A hybrid meshing framework adapted to the topography to simulate atmospheric boundary layer flows

  • Gargallo Peiró, Abel|||0000-0003-3742-2197
  • Avila, Matias
  • Folch, Arnau|||0000-0002-0677-6366
A new topography adapted mesh generation framework tailored to simulate Atmospheric Boundary Layer (ABL) flows on complex terrains is presented. The mesher is fully automatic given: the maximum and minimum surface mesh size, and the mesh size at the top of the ABL. The following contributions to the meshing workflow for ABL flow simulation are performed. First, we present a smooth topography modeling to query first and second-order geometry derivatives. Second, we propose a new adaptive meshing procedure to discretize the topography based on two different metrics. Third, we present the ABL mesher, featuring both prisms and tetrahedra. We extrude the triangles of the adapted surface mesh, generating prisms that reproduce the Surface Boundary Layer. Then, the rest of the domain is meshed with an unstructured tetrahedral mesh. In addition, we detail a hybrid quality optimization approach for both the surface and volume meshers, analyzing its impact on the solver for high-complexity terrains. We analyze the convergence of the triangle adaptive approach, obtaining quadratic convergence to the geometry and reducing to one-half the error for the same amount of degrees of freedom than without adaptivity and optimization. We also study the mesh convergence of our Reynolds-averaged Navier–Stokes (RANS) solver, obtaining quadratic mesh convergence to the solution, and using a 30% of the degrees of freedom while reducing a 20% of the error of standard semi-structured approaches. Finally, we present the generated meshes and the simulation results for a complete complex topographic scenario., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Optimization strategies for geophysics models on manycore systems

  • Serpa, Matheus S.
  • Cruz, Eduardo HM
  • Diener, Matthias
  • Krause, Arthur M.
  • Navaux, Philippe
  • Panetta, Jairo
  • Farrés Coma, Albert
  • Rosas, Claudia
  • Hanzich, Mauricio
Many software mechanisms for geophysics exploration in oil and gas industries are based on wave propagation simulation. To perform such simulations, state-of-the-art high-performance computing architectures are employed, generating results faster with more accuracy at each generation. The software must evolve to support the new features of each design to keep performance scaling. Furthermore, it is important to understand the impact of each change applied to the software to improve the performance as most as possible. In this article, we propose several optimization strategies for a wave propagation model for six architectures: Intel Broadwell, Intel Haswell, Intel Knights Landing, Intel Knights Corner, NVIDIA Pascal, and NVIDIA Kepler. We focus on improving the cache memory usage, vectorization, load balancing, portability, and locality in the memory hierarchy. We analyze the hardware impact of the optimizations, providing insights of how each strategy can improve the performance. The results show that NVIDIA Pascal outperforms the other considered architectures by up to 8.5×., Our research received funding from the EU H2020 Programme and from MCTI/RNP-Brazil under the HPC4E project, grant agreement 689772, as well as from CNPq and Capes., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

A hybrid meshing framework adapted to the topography to simulate atmospheric boundary layer flows

  • Gargallo Peiró, Abel
  • Avila, Matias
  • Folch, Arnau
A new topography adapted mesh generation framework tailored to simulate Atmospheric Boundary Layer (ABL) flows on complex terrains is presented. The mesher is fully automatic given: the maximum and minimum surface mesh size, and the mesh size at the top of the ABL. The following contributions to the meshing workflow for ABL flow simulation are performed. First, we present a smooth topography modeling to query first and second-order geometry derivatives. Second, we propose a new adaptive meshing procedure to discretize the topography based on two different metrics. Third, we present the ABL mesher, featuring both prisms and tetrahedra. We extrude the triangles of the adapted surface mesh, generating prisms that reproduce the Surface Boundary Layer. Then, the rest of the domain is meshed with an unstructured tetrahedral mesh. In addition, we detail a hybrid quality optimization approach for both the surface and volume meshers, analyzing its impact on the solver for high-complexity terrains. We analyze the convergence of the triangle adaptive approach, obtaining quadratic convergence to the geometry and reducing to one-half the error for the same amount of degrees of freedom than without adaptivity and optimization. We also study the mesh convergence of our Reynolds-averaged Navier–Stokes (RANS) solver, obtaining quadratic mesh convergence to the solution, and using a 30% of the degrees of freedom while reducing a 20% of the error of standard semi-structured approaches. Finally, we present the generated meshes and the simulation results for a complete complex topographic scenario., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Hardware-Assisted Thread and Data Mapping in Hierarchical Multicore Architectures

  • Cruz, Eduardo H.M.
  • Diener, Matthias
  • Pilla, Laércio L.
  • Navaux, Philippe O.A.
The performance and energy efficiency of modern architectures depend on memory locality, which can be improved by thread and data mappings considering the memory access behavior of parallel applications. In this article, we propose intense pages mapping, a mechanism that analyzes the memory access behavior using information about the time the entry of each page resides in the translation lookaside buffer. It provides accurate information with a very low overhead. We present experimental results with simulation and real machines, with average performance improvements of 13.7% and energy savings of 4.4%, which come from reductions in cache misses and interconnection traffic., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

The combustion mitigation of methane as a non-CO2 greenhouse gas

  • Jian, X.
  • Mira Martínez, Daniel
  • Cluff, D.L.
Anthropogenic emissions of non-CO2 greenhouse gases such as fugitive methane contribute significantly to global warming. A review of fugitive methane combustion mitigation and utilisation technologies, which are primarily aimed at methane emissions from coal mining activities, with a focus on modelling and simulation of ultra-lean methane oxidation/combustion is presented. The challenges associated with ultra-lean methane oxidation are on the ignition of the ultra-lean mixture and sustainability of the combustion process. There is a lack of fundamental studies on chemical kinetics of ultra-lean methane combustion and reliable kinetic schemes that can be used together with computational fluid dynamics studies to design and develop advanced mitigation systems. Mitigation of methane as a greenhouse gas calls for more efforts on understanding ultra-lean combustion. Recuperative combustion provides a promising means for mitigating ultra-lean methane emissions. Progress is needed on effective methods to ignite and to recuperate and retain heat for oxidation/combustion of the ultra-lean mixtures. Catalysts can be very effective in reducing the temperatures required for oxidation while plasmas may be utilised to assist the ignition, but thermodynamic/aerodynamic limits of burning ultra-lean methane remain unexplored. Further technological developments may be focussed on developing innovative capturing technology as well as technological innovations to achieve effective ignition and sustainable oxidation/combustion., We would like to thank the reviewers and the editor for their constructive suggestions on the manuscript. These research results have received funding from the EU H2020 Programme (No. 689772) and from MCTI/RNP-Brazil under the HPC4E Project, grant agree- ment no 689772., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Parallel mesh partitioning based on space filling curves

  • Borrell, Ricard
  • Cajas García, Juan Carlos
  • Mira Martínez, Daniel
  • Taha, A.
  • Koric, S.
  • Vázquez, Mariano
  • Houzeaux, Guillaume
Larger supercomputers allow the simulation of more complex phenomena with increased accuracy. Eventually this requires finer and thus also larger geometric discretizations. In this context, and extrapolating to the Exascale paradigm, meshing operations such as generation, deformation, adaptation/regeneration or partition/load balance, become a critical issue within the simulation workflow. In this paper we focus on mesh partitioning. In particular, we present a fast and scalable geometric partitioner based on Space Filling Curves (SFC), as an alternative to the standard graph partitioning approach. We have avoided any computing or memory bottleneck in the algorithm, while we have imposed that the solution achieved is independent (discounting rounding off errors) of the number of parallel processes used to compute it. The performance of the SFC-based partitioner presented has been demonstrated using up to 4096 CPU-cores in the Blue Waters supercomputer., The research leading to these results has received funding from the European Union Horizon 2020 Programme (2014-2020) and the Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (grant agreement No. 689772). This work is also part of the PRAC "Simulations of Aircraft Engines" supported by the National Science Foundation. It has also been financially supported by the PRACE preparatory access projects funded in part by the EU Horizon 2020 research and innovation programme (2014-2020) under grant agreement 653838. J.C. Cajas acknowledges the nancial sup- port of the `Consejo Nacional de Ciencia y Tecnolog a (CONACyT, M exico)' grant number 231588 290790. Ricard Borrell and Daniel Mira acknowledge the Juan de la Cierva postdoctoral grants with codes IJCI-2014-21034 and IJCI-2015-26686, respectively., Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Wavefield compression for adjoint methods in full-waveform inversion

  • Boehm, Christian
  • Hanzich, Mauricio
  • de la Puente, Josep
  • Fichtner, Andreas
Adjoint methods are a key ingredient of gradient-based full-waveform inversion schemes. While being conceptually elegant, they face the challenge of massive memory requirements caused by the opposite time directions of forward and adjoint simulations and the necessity to access both wavefields simultaneously for the computation of the sensitivity kernel. To overcome this bottleneck, we have developed lossy compression techniques that significantly reduce the memory requirements with only a small computational overhead. Our approach is tailored to adjoint methods and uses the fact that the computation of a sufficiently accurate sensitivity kernel does not require the fully resolved forward wavefield. The collection of methods comprises reinterpolation with a coarse temporal grid as well as adaptively chosen polynomial degree and floating-point precision to represent spatial snapshots of the forward wavefield on hierarchical grids. Furthermore, the first arrivals of adjoint waves are used to identify “shadow zones” that do not contribute to the sensitivity kernel. Numerical experiments show the high potential of this approach achieving an effective compression factor of three orders of magnitude with only a minor reduction in the rate of convergence. Moreover, it is computationally cheap and straightforward to integrate in finite-element wave propagation codes with possible extensions to finite-difference methods., The authors would like to thank Editor John Etgen, Associate Editor Anatoly Baumstein and three anonymous reviewers for their valuable comments and excellent suggestions to improve the paper. We gratefully acknowledge support by the Swiss National Supercomputing Centre (CSCS) in the form of CHRONOS project ch1 and PASC project GeoScale. The research leading to these results has received funding from the European Union's Horizon 2020 Programme (2014-2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (www.hpc4e.eu), grant agreement n. 689772. Furthermore, the first author acknowledges funding from Shell within the project \Boosting full-waveform inversion"., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

A CFD framework for offshore and onshore wind farm simulation

  • Avila, Matias
  • Gargallo-Peiró, Abel
  • Folch, Arnau
We present a wind simulation framework for offshore and onshore wind farms. The simulation framework involves an automatic hybrid high-quality mesh generation process, a pre-processing to impose initial and boundary conditions, and a solver for the Reynolds Averaged Navier-Stokes (RANS) equations with two different turbulence models, a modified standard k-epsilon model and a realizable k-epsilon model in which we included the Coriolis effects. Wind turbines are modeled as actuator discs. The wind farm simulation framework has been implemented in Alya, an in-house High Performance Computing (HPC) multi-physics finite element parallel solver. An application example is shown for an onshore wind farm composed of 165 turbines., This work has been supported by the EU H2020 projects New European Wind Atlas ERA-NET PLUS (NEWA), High Performance Computing for Energy (HPC4E, grant agreement 689772), and the Energy oriented Centre of Excellence (EoCoE, grant agreement 676629). We also thank two anonymous reviewers for their constructive comments on the first version of the manuscript., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

MPI+X: task-based parallelization and dynamic load balance of finite element assembly

  • Garcia-Gasulla, Marta
  • Houzeaux, Guillaume
  • Ferrer, Roger
  • Artigues, Antoni
  • López, Victor
  • Labarta Mancho, Jesús José
  • Vázquez, Mariano
The main computing tasks of a finite element code(FE) for solving partial differential equations (PDE's) are the algebraic system assembly and the iterative solver. This work focuses on the first task, in the context of a hybrid MPI+X paradigm. Although we will describe algorithms in the FE context, a similar strategy can be straightforwardly applied to other discretization methods, like the finite volume method. The matrix assembly consists of a loop over the elements of the MPI partition to compute element matrices and right-hand sides and their assemblies in the local system to each MPI partition. In a MPI+X hybrid parallelism context, X has consisted traditionally of loop parallelism using OpenMP. Several strate- gies have been proposed in the literature to implement this loop parallelism, like coloring or substructuring techniques to circumvent the race condition that appears when assembling the element system into the local system. The main drawback of the first technique is the decrease of the IPC due to bad spatial locality. The second technique avoids this issue but requires extensive changes in the implementation, which can be cumbersome when several element loops should be treated. We propose an alternative, based on the task parallelism of the element loop using some extensions to the OpenMP programming model. The task- ification of the assembly solves both aforementioned problems. In addition, dynamic load balance will be applied using the DLB library, especially efficient in the presence of hybrid meshes, where the relative costs of the different elements is impossible to estimate a priori. This paper presents the proposed methodology, its implementation and its validation through the solution of large computational mechanics problems up to 16k cores., The use of large part of a supercomputer, even more in normal conditions of use, is never an innocuous exercise. The research leading to these results has received funding from: the European Union's Horizon 2020 Programme (2014–2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP), HPC4E Project, grant agreement 689772; the Energy oriented Centre of Excellence (EoCoE), grant agreement number 676629, funded within the Horizon2020 framework of the European Union; The Spanish Government (grant SEV2015-0493 of the Severo Ochoa Program); the Spanish Ministry of Science and Innovation (contract TIN2015-65316-P); the Generalitat de Catalunya (contract 2014-SGR-1051); the Intel-BSC Exascale Lab collaboration project. Comissió Interdepartamental de Recerca i Innovació Tecnológica(Interdepartmental Commission for Research and Technological Innovation), Sí, Post-print (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

MPI+X: task-based parallelisation and dynamic load balance of finite element assembly

  • Garcia, Marta
  • Houzeaux, Guillaume
  • Ferrer, Roger
  • Artigues, Antoni
  • López, Victor
  • Labarta Mancho, Jesús José
  • Vázquez, Mariano
The main computing phases of numerical methods for solving partial differential equations are the algebraic system assembly and the iterative solver. This work focuses on the first task, in the context of a hybrid MPI+X paradigm. The matrix assembly consists of a loop over the elements, faces, edges or nodes of the MPI partitions to compute element matrices and vectors and then of their assemblies. In a MPI+X hybrid parallelism context, X has consisted traditionally of loop parallelism using OpenMP, with different techniques to avoid the race condition, but presenting efficiency or implementation drawbacks. We propose an alternative, based on task parallelism using some extensions to the OpenMP programming model. In addition, dynamic load balance will be applied, especially efficient in the presence of hybrid meshes. This paper presents the proposed methodology, its implementation and its validation through the solution of large computational mechanics problems up to 16k cores., The research leading to these results has received funding from: the European Union's Horizon 2020 Programme (2014–2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP), HPC4E Project, grant agreement 689772; the Energy oriented Centre of Excellence (EoCoE), grant agreement number 676629, funded within the Horizon2020 framework of the European Union; The Spanish Government (grant SEV2015-0493 of the Severo Ochoa Program); the Spanish Ministry of Science and Innovation (contract TIN2015-65316-P); the Generalitat de Catalunya (contract 2014-SGR-1051); the Intel-BSC Exascale Lab collaboration project. Comissió Interdepartamental de Recerca i Innovació Tecnológica(Interdepartmental Commission for Research and Technological Innovation), Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Toward an automatic full-wave inversion: Synthetic study cases

  • Kormann, Jean
  • Rodriguez, Juan E.
  • Gutiérrez, Natalia
  • Ferrer, Miguel
  • Rojas, Otilio
  • de la Puente, Josep
  • Hanzich, Mauricio
  • Cela, Jose M.
Full-waveform inversion (FWI) in seismic scenarios continues to be a complex procedure for subsurface imaging that might require extensive human interaction in terms of model setup, constraints, and data preconditioning. The underlying reason is the strong nonlinearity of the problem that forces the addition of a priori knowledge (or bias) in order to obtain geologically sound results. In particular, when the use of a long-offset receiver is not possible or may not favor the reconstruction of the fine structure of the model, one needs to rely on reflection data. As a consequence, the inversion process is more prone to becoming stuck in local minima. Nevertheless, misfit functionals can be devised that can either cope with missing long-wavenumber features of initial models (e.g., cross-correlation-based misfit) or invert reflection-dominated data whenever the models are sufficiently good (e.g., normalized offset-limited least-squares misfit). By combining both, high-frequency data content with poor initial models can be successfully inverted. If one can figure out simple parameterizations for such functionals, the amount of uncertainty and manual work related to tuning FWI would be substantially reduced. Thus, FWI might become a semiautomatized imaging tool., We want to thank Repsol for funding this research by means of the Aurora project. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 644202. Additionally, the research leading to these results has received funding from the European Union’s Horizon 2020 Programme (2014-2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (www.hpc4e.eu), grant agreement No 689772. We acknowledge Chevron for the dataset that was used in our second example., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Diurnal cycle RANS simulations applied to wind resource assessment

  • Barcons, Jordi
  • Avila, Matias
  • Folch, Arnau
Microscale computational fluid dynamics (CFD) models can be used for wind resource assessment on complex terrains. These models generally assume neutral atmospheric stratification, an assumption that can lead to inaccurate modeling results and to large uncertainties at certain sites. We propose a methodology for wind resource evaluation based on unsteady Reynolds averaged Navier‐Stokes (URANS) simulations of diurnal cycles including the effect of thermal stratification. Time‐dependent boundary conditions are generated by a 1D precursor to drive 3D diurnal cycle simulations for a given geostrophic wind direction sector. Time instants of the cycle representative of four thermal stability regimes are sampled within diurnal cycle simulations and combined with masts time series to obtain the wind power density (WPD). The methodology has been validated on a complex site instrumented with seven met masts. The WPD spatial distribution is in good agreement with observations with the mean absolute error improving 17.1% with respect to the neutral stratification assumption., This work has been partially supported by the three EU H2020 projects, New European Wind Atlas ERA‐NET PLUS (NEWA, FP7‐ENERGY.2013.10.1.2, European Commission's grant agreement 618122), High Performance Computing for Energy (HPC4E, grant agreement 689772), and the Energy oriented Centre of Excellence (EoCoE, grant agreement 676629), and the SEDAR (“Simulación eólica de alta resolución”) project. Jordi Barcons is grateful to a PhD fellowship from the Industrial Doctorates Plan of the Government of Catalonia (Ref. eco/2497/2013). We also thank Iberdrola Renovables Energa S.A. and Impulsora Latinoamericana de Energa Renovables S.A. for providing the access to Puebla met masts data for validation and to Luis Prieto and Daniel Paredes for their help. We also thank the reviewers for their productive comments and observations., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Heat loss prediction of a confined premixed jet flame using a conjugate heat transfer approach

  • Gövert, S.
  • Mira Martínez, Daniel
  • Zavala-Ake, M.
  • Kok, J.B.W.
  • Vázquez, Mariano
  • Houzeaux, Guillaume
The presented work addresses the investigation of the heat loss of a confined turbulent jet flame in a lab-scale combustor using a conjugate-heat transfer approach and large-eddy simulation. The analysis includes the assessment of the principal mechanisms of heat transfer in this combustion chamber: radiation, convection and conduction of heat over walls. A staggered approach is used to couple the reactive flow field to the heat conduction through the solid and both domains are solved using two implementations of the same code. Numerical results are compared against experimental data and an assessment of thermal boundary conditions to improve the prediction of the reactive flow field is given., The research leading to these results has received funding through the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (FP7, 2007–2013) under the Grant agreement No. FP7-290042 for the project COPA-GT as well as the European Union’s Horizon 2020 Programme (2014–2020) and from Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project, Grant agreement No. 689772. The authors thankfully acknowledge the computer resources, technical expertise and assistance provided by the Red Española de Supercomputación (RES). Finally, the authors would like to thank O. Lammel for the useful discussions and kindly providing the data for the comparison., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

A wind field downscaling strategy based on domain segmentation and transfer functions

  • Barcons, Jordi
  • Avila, Matias
  • Folch, Arnau
This paper presents a novel methodology for mesoscale‐to‐microscale downscaling of near‐surface wind fields. The model chain consists on the Weather Research and Forecast mesoscale model and the Alya‐CFDWind microscale model (assuming neutral stability). The downscaling methodology combines precomputed microscale simulations with a mesoscale forecast using a domain segmentation technique and transfer functions. As a result, the downscaled wind field preserves the mesoscale pattern but, at the same time, incorporates local mesoscale subgrid terrain effects, particularly at valleys and channelling zones. The methodology has been validated over a 9‐month period on a very complex terrain site instrumented with a dense observational network of meteorological masts. With respect to mesoscale results, the global skills of the downscaled wind at masts improve for wind direction and remain similar for wind velocity. However, a substantial improvement occurs under stable and neutral conditions and for high wind velocity regimes., This work has been partially funded by the High Performance Computing for Energy (HPC4E) project (call H2020-EUB-2015, Topic: EUB-2-2015, type of action RIA, Grant Agreement number 689772) and the SEDAR ("Simulación eólica de alta resolución") project. It has also been partially supported by the Energy-oriented Centre of Excellence (EoCoE) (Grant Agreement number 676629, funded within the H2020 framework of the EuropeanUnion). J.B. is grateful to a PhD fellowship from the IndustrialDoctorates Plan of the Government of Catalonia (Ref. eco/2497/2013). We also thank Daniel Paredes and Luis Prieto from Iberdrola Renovables S.A. for providing us access to met masts data for validation., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Multi-objective scheduling of Scientific Workflows in multisite clouds

  • Liu, Ji
  • Pacitti, Esther
  • Valduriez, Patrick
  • Oliveira, Daniel de
  • Mattoso, Marta
Clouds appear as appropriate infrastructures for executing Scientific Workflows (SWfs). A cloud is typically made of several sites (or data centers), each with its own resources and data. Thus, it becomes important to be able to execute some SWfs at more than one cloud site because of the geographical distribution of data or available resources among different cloud sites. Therefore, a major problem is how to execute a SWf in a multisite cloud, while reducing execution time and monetary costs. In this paper, we propose a general solution based on multi-objective scheduling in order to execute SWfs in a multisite cloud. The solution consists of a multi-objective cost model including execution time and monetary costs, a Single Site Virtual Machine (VM) Provisioning approach (SSVP) and ActGreedy, a multisite scheduling approach. We present an experimental evaluation, based on the execution of the SciEvol SWf in Microsoft Azure cloud. The results reveal that our scheduling approach significantly outperforms two adapted baseline algorithms (which we propose by adapting two existing algorithms) and the scheduling time is reasonable compared with genetic and brute-force algorithms. The results also show that our cost model is accurate and that SSVP can generate better VM provisioning plans compared with an existing approach., Work partially funded by EU H2020 Programme and MCTI/RNP-Brazil (HPC4E grant agreement number 689772), CNPq, FAPERJ, and INRIA (MUSIC project), Microsoft (ZcloudFlow project) and performed in the context of the Computational Biology Institute (www.ibc-montpellier.fr). We would like to thank Kary Ocaña for her help in modeling and executing the SciEvol SWf., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

The Effect of Partial Premixing and Heat Loss on the Reacting Flow Field Prediction of a Swirl Stabilized Gas Turbine Model Combustor

  • Gövert, Simon
  • Mira Martínez, Daniel
  • Kok, Jim B. W.
  • Vázquez, Mariano
  • Houzeaux, Guillaume
This work addresses the prediction of the reacting flow field in a swirl stabilized gas turbine model combustor using large-eddy simulation. The modeling of the combustion chemistry is based on laminar premixed flamelets and the effect of turbulence-chemistry interaction is considered by a presumed shape probability density function. The prediction capabilities of the presented combustion model for perfectly premixed and partially premixed conditions are demonstrated. The effect of partial premixing for the prediction of the reacting flow field is assessed by comparison of a perfectly premixed and partially premixed simulation. Even though significant mixture fraction fluctuations are observed, only small impact of the non-perfect premixing is found on the flow field and flame dynamics. Subsequently, the effect of heat loss to the walls is assessed assuming perfectly premixing. The adiabatic baseline case is compared to heat loss simulations with adiabatic and non-adiabatic chemistry tabulation. The results highlight the importance of considering the effect of heat loss on the chemical kinetics for an accurate prediction of the flow features. Both heat loss simulations significantly improve the temperature prediction, but the non-adiabatic chemistry tabulation is required to accurately capture the chemical composition in the reacting layers., The research leading to these results has received funding through the People Program (Marie Curie Actions) of the European Union’s Seventh Framework Program (FP7, 2007-2013) for the project COPA-GT (grant agreement No. FP7-290042) as well as the European Union’s Horizon 2020 Program (2014-2020) and the Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (grant agreement No. 689772). Furthermore, computer resources and technical assistance has been provided by the Red Española de Supercomputación (RES) (grant numbers FI-2015-1-0002, FI-2015-3-0013, FI-2016-1-0001)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Mesh generation, sizing and convergence for onshore and offshore wind farm Atmospheric Boundary Layer flow simulation with actuator discs

  • Gargallo Peiró, Abel
  • Avila, Matias
  • Owen, Herbert
  • Prieto-Godino, Luis
  • Folch, Arnau
A new mesh generation process for wind farm modeling is presented together with a mesh convergence and sizing analysis for wind farm flow simulations. The generated meshes are tailored to simulate Atmospheric Boundary Layer (ABL) flows on complex terrains modeling the wind turbines as actuator discs. The wind farm mesher is fully automatic and, given the topography and the turbine characteristics (location, diameter and hub height), it generates a hybrid mesh conformal with the actuator discs and refined upwind and downstream. Moreover, it presents smooth element size transitions across scales and avoids extending high-resolution areas to all the domain. We take advantage of our automatic and robust mesher to study the mesh convergence of our RANS solver with linear elements, obtaining quadratic mesh convergence for a quantity of interest in all the tested cases. In addition, we quantify the mesh resolution at the terrain surface and at the actuator discs required to achieve a given numerical error in simulations in onshore and offshore frameworks. Finally, we present the generated meshes and the simulation results for an offshore and an onshore wind farm. We analyze in detail one particular wind direction for both cases, and for the onshore wind farm we use our automatic framework to estimate the yearly production of energy and measuring the error against the actual produced one., This work has been partially funded by the EU H2020Energy oriented Center of Excellence (EoCoE) for computer applications, the New European Wind Atlas (NEWA) and the High Performance Computing for Energy (HPC4E) projects. We thank Iberdrola Renovables for their collaboration and for providing wind farm data to validate the developed techniques., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Acceleration strategies for elastic full waveform inversion workflows in 2D and 3D

  • Kormann, Jean
  • Rodríguez, Juan E.
  • Ferrer, Miguel
  • Farrés, Albert
  • Gutiérrez, Natalia
  • de la Puente, Josep
  • Hanzich, Mauricio
  • Cela, José M.
Full waveform inversion (FWI) is one of the most challenging procedures to obtain quantitative information of the subsurface. For elastic inversions, when both compressional and shear velocities have to be inverted, the algorithmic issue becomes also a computational challenge due to the high cost related to modelling elastic rather than acoustic waves. This shortcoming has been moderately mitigated by using high-performance computing to accelerate 3D elastic FWI kernels. Nevertheless, there is room in the FWI workflows for obtaining large speedups at the cost of proper grid pre-processing and data decimation techniques. In the present work, we show how by making full use of frequency-adapted grids, composite shot lists and a novel dynamic offset control strategy, we can reduce by several orders of magnitude the compute time while improving the convergence of the method in the studied cases, regardless of the forward and adjoint compute kernels used., The authors thank REPSOL for the permission to publish the present research and for funding through the AURORA project. J. Kormann also thankfully acknowledges the computer resources, technical expertise and assistance provided by the Barcelona Supercomputing Center - Centro Nacional de Supercomputacti ´on together with the Spanish Supercomputing Network (RES) through grant FI-2014-2-0009. This project has received funding from the European Union’s Horizon 2020, research and innovation programme under the Marie Skłodowska-Curie grant agreement no. 644202. The research leading to these results has received funding from the European Union’s Horizon 2020 Programme (2014–2020) and from the Brazilian Ministry of Science, Technology and Innovation through Rede Nacional de Pesquisa (RNP) under the HPC4E Project (www.hpc4e.eu), grant agreement no. 689772.We further want to thank the Editor Clint N. Dawson for his help, and Andreas Fichtner and an anonymous reviewer for their comments and suggestions to improve the manuscript., Peer Reviewed, Postprint (published version)