International Conference on Computacional Methods for Coupled Problems in Science and Engineering

p. 1

Presentation's date: 2015

Abstract:

The simulation of Fluid Structure Interaction (FSI) problems requires by its very nature taking into account relevant displacements of solids within a fluid domain. While “small” deformations can be conveniently represented by Arbitrary Lagrangian Eulerian (ALE) techniques, such approaches fail when facing large deformations. One possible solution is the use of “Embedded Solvers” which essentially consist in embedding an approximation of the geometry within the fluid discretization so that such information can be taken into account while solving the fluid problem. Current work focuses on the implementation of one of such Embedded approaches in the context of Variational Multi Scale (VMS) techniques. We will show how this technique can be employed to perform effectively fully coupled LES-like simulations. The method is completed by the use of the parallel adaptive refinement strategy described in [1]. The method is evaluated in application to a real deformable structure for which experimental results are available.]]>

Abstract:

The importance of advanced simulation to the competitiveness of both large and small companies is well established. The principal objective of Fortissimo is to enable European manufacturing, particularly small to medium enterprises (SMEs), to benefit from the efficiency and competitive advantage inherent in the use of simulation. However, the simulation of, for example, high-pressure gas cylinders, the moulding of plastics or the thermodynamic properties of hazardous materials requires enormous computing power and specialised software tools and services. Generally, large companies, which have a greater pool of skills and resources, find access to advanced simulation easier than SMEs which can neither afford expensive High Performance Computing equipment nor the licensing cost for the relevant tools. This means that SMEs are not able to take advantage of advanced simulation, even though it can clearly make them more competitive. The goal of Fortissimo is to overcome this impasse through the provision of simulation services running on a cloud infrastructure making use of High Performance Computing systems also making appropriate skills and tools available in a distributed, internet-based environment.\n\nFortissimo will make advanced simulation more easily accessible, particularly to SMEs, through the realisation of a 'one-stop shop' where hardware, expertise, applications, visualisation and tools will be easily available and affordable on a pay-per-use basis. In doing this it will create and demonstrate a sustainable commercial ecosystem where actors at all levels in the value chain can realise sufficient commercial benefit to enable that ecosystem to persist independently of EU funding and continue to provide affordable services to manufacturing industry, particularly SMEs.\n\nFortissimo will be driven by end-user requirements where (~50) business-relevant application experiments will be used to develop, test and demonstrate both the infrastructure and the 'one-stop pay-per-use shop'. The project participants represent all actors in the value chain. Not only will Fortissimo contribute to the increased competitiveness of European manufacturing industry through the innovative infrastructure that it will develop and test, but it will create commercial opportunities for European Independent Software Vendors, as well as for service and High Performance Computing infrastructure providers, through the creation of a new market for their products and services. Fortissimo places considerable emphasis on the exploitation of opportunities at all levels of the value chain ranging from the end-user to the High Performance Computing infrastructure provider.\n\nFortissimo involves 1,132 months of effort, a total cost of €21.7m and EC funding of €16m over a duration of three years, commensurate with achieving its ambitious goals.]]>

Computers and fluids

Vol. 80, num. 1, p. 301-309

DOI: 10.1016/j.compfluid.2012.02.004

Date of publication: 2013-07-10

Abstract:

Creating a highly parallelizable code is a challenge specially for Distributed Memory Machines (DMMs). Moreover, algorithms and data structures suitable for these platforms can be very different from the ones used in serial code. For this reason, many programmers in the field prefer to start their own code from scratch. However, for an already existing framework supported by a long-time expertise the idea of transformation becomes attractive in order to reuse the effort done during years of development. In this presentation we explain how a relatively complex framework but with modular structure can be prepared for high performance computing with minimum modification. Kratos Multi-Physics [1] is an open source generic multi-disciplinary platform for solution of coupled problems consist of fluid, structure, thermal and electromagnetic fields. The parallelization of this framework is performed with objective of enforcing the less possible changes to its different solver modules and encapsulate the changes as much as possible in its common kernel. This objective is achieved thanks to the Kratos design and also innovative way of dealing with data transfers for a multi-disciplinary code. This work is completed by the migration of the framework from the 86× architecture to the Marenostrum Supercomputing platform. The migration has been verified by a set of benchmarks which show high scalability, from which we present the Telescope problem in this paper.

Creating a highly parallelizable code is a challenge specially for distributed memory machines (DMMs). Moreover, algorithms and data structures suitable for these platforms can be very different from the ones used in serial code. For this reason, many programmers in the field prefer to start their own code from scratch. However, for an already existing framework supported by a long-time expertise the idea of transformation becomes attractive in order to reuse the effort done during years of development. In this presentation we explain how a relatively complex framework but with modular structure can be prepared for high performance computing with minimum modification. Kratos Multi-Physics [1] is an open source generic multi-disciplinary platform for solution of coupled problems consist of fluid, structure, thermal and electromagnetic fields. The parallelization of this framework is performed with objective of enforcing the less possible changes to its different solver modules and encapsulate the changes as much as possible in its common kernel. This objective is achieved thanks to the Kratos design and also innovative way of dealing with data transfers for a multi-disciplinary code. This work is completed by the migration of the framework from the x86 architecture to the Marenostrum Supercomputing platform. The migration has been verified by a set of benchmarks which show high scalability, from which we present the Telescope problem in this paper.]]>

Computers and fluids

Vol. 80, p. 342-355

DOI: 10.1016/j.compfluid.2012.01.023

Date of publication: 2013-07

Abstract:

The present article describes a simple element-driven strategy for the conforming refinement of simplicial finite element meshes in a distributed environment. The proposed algorithm is effective both for local adaptive refinement and for the division of all the elements within an existing mesh. We aim to provide sufficient detail to allow the practical implementation of the algorithm, which can be coded with minimal effort provided that a distributed linear algebra library is available. The proposed refinement strategy is composed of three basic components: a global splitting strategy, an elemental splitting procedure and an error estimation technique, which are combined so to guarantee obtaining a conformant refined mesh. A number of benchmark examples show the capabilities of the proposed method. Error is estimated for the incompressible fluid-flow benchmarks using a novel indicator based on the computation of the sub-scale velocity.

The present article describes a simple element-driven strategy for the conforming refinement of simplicial finite element meshes in a distributed environment. The proposed algorithm is effective both for local adaptive refinement and for the division of all the elements within an existing mesh. We aim to provide sufficient detail to allow the practical implementation of the algorithm, which can be coded with minimal effort provided that a distributed linear algebra library is available. The proposed refinement strategy is composed of three basic components: a global splitting strategy, an elemental splitting procedure and an error estimation technique, which are combined so to guarantee obtaining a conformant refined mesh. A number of benchmark examples show the capabilities of the proposed method. Error is estimated for the incompressible fluid-flow benchmarks using a novel indicator based on the computation of the sub-scale velocity.]]>

European Congress on Computational Methods in Applied Sciences and Engineering

Presentation's date: 2012-09-12

International Conference on Parallel Computational Fluid Dynamics

p. 1-5

Presentation's date: 2011-05-17

Abstract:

Creating a highly parallelizable code is a challenge and development for distributed memory machines (DMMs) can be very different form developing a serial code in term of algorithms and structure. For this reason, many developers in the field prefer to develop their own code from scratch. However, for an already existing framework with large development background the idea of transformation becomes attractive in order to reuse the effort done during years of development. In this presentation we explain how a relatively complex framework but with modular structure can be prepared for high performance computing with minimum modification. Kratos Multi-Physics [1] is an open source generic multi-disciplinary platform for solution of coupled problems consist of fluid, structure, thermal and electromagnetic fields. The parallelization of this framework is performed with objective of enforcing the less possible changes to its different solver modules and encapsulate the changes as much as possible in its common kernel. This objective is achieved thanks to the Kratos design and also innovative way of dealing with data transfers for a multi-disciplinary code. This work is completed by the migration of the framework from the x86 architecture to the Marenostrum Supercomputing platform. The migration has been verified by a set of benchmarks which show very good scalability, from which we present the Telescope problem in this paper.]]>

International Conference on Parallel Computational Fluid Dynamics

p. 1-5

Presentation's date: 2011-05

Abstract:

Dealing with large simulation is a growing challenge. Ideally for the wellparallelized software prepared for high performance, the problem solving capability depends on the available hardware resources. But in practice there are several technical details which reduce the scalability of the system and prevent the effective use of such a software for large problems. In this work we describe solutions implemented in order to obtain a scalable system to solve and visualize large scale problems. The present work is based on Kratos MutliPhysics [1] framework in combination with GiD [2] pre and post processor. The applied techniques are verified by CFD simulation and visualization of a wind tunnel problem with more than 100 millions of elements in our in-hose cluster in CIMNE.

Dealing with large simulation is a growing challenge. Ideally for the wellparallelized software prepared for high performance, the problem solving capability depends on the available hardware resources. But in practice there are several technical details which reduce the scalability of the system and prevent the effective use of such a software for large problems. In this work we describe solutions implemented in order to obtain a scalable system to solve and visualize large scale problems. The present work is based on Kratos MutliPhysics [1] framework in combination with GiD [2] pre and post processor. The applied techniques are verified by CFD simulation and visualization of a wind tunnel problem with more than 100 millions of elements in our in-hose cluster in CIMNE.]]>

International Conference on Parallel, Distributed, Grid and Cloud Computing for Engineering

DOI: 10.4203/ccp.95.56

Presentation's date: 2011-04

Abstract:

In this paper, we implement a parallel solver for the incompressible Navier-Stokes equations using the finite element method. We compare two different parallel programming strategies, OpenMP, which is based on a shared memory model, and MPI, which uses a distributed memory model. The incompressible Navier-Stokes equations constitute a transient, non-linear system which has well-known stability issues when the standard Galerkin finite element discretization is used. This means that some treatment will be required before we can implement their solution. We implement a solution strategy based on algebraic subgrid scale stabilization and a generalized Newmark method for iteration in time. This results in a linear system which, unfortunately, can be poorly scaled, as it involves both velocity and pressure degrees of freedom. As a result, the system is difficult to solve using iterative methods, which are preferable for large problems. To avoid this situation, we have used an uncoupling approach based on a Schur complement formulation for the pressure, which results in a scheme similar to that of fractional step methods. Two different implementations of this method, one using OpenMP and another based on MPI through the Trilinos library, have been used to solve Ahmed's body, a standard benchmark problem in turbulence which simulates the flow around a bus-like object. Both implementations are compared with each other and with a scalar run of the same example, with emphasis on performance gains provided by the parallelization. In this sense, we have found that the time benefit obtained from the parallelization is affected by the hardware where the simulations are run. In our case, this imposes us two clear restrictions: OpenMP is limited to processors that can access some shared memory, and in general the performance is affected by competition between the processes for memory access, which seems to be an important bottleneck in our test system. In general, it can be said that, when choosing an approach to parallelization, it is important to take into account the available hardware.]]>