Filters








1,758 Hits in 1.1 sec

Parallel Machine Learning of Partial Differential Equations [article]

Amin Totounferoush, Neda Ebrahimi Pour, Sabine Roller, Miriam Mehl
2021 arXiv   pre-print
In this work, we present a parallel scheme for machine learning of partial differential equations. The scheme is based on the decomposition of the training data corresponding to spatial subdomains, where an individual neural network is assigned to each data subset. Message Passing Interface (MPI) is used for parallelization and data communication. We use convolutional neural network layers (CNN) to account for spatial connectivity. We showcase the learning of the linearized Euler equations to
more » ... sess the accuracy of the predictions and the efficiency of the proposed scheme. These equations are of particular interest for aeroacoustic problems. A first investigation demonstrated a very good agreement of the predicted results with the simulation results. In addition, we observe an excellent reduction of the training time compared to the sequential version, providing an almost perfect scalability up to 64 CPU cores.
arXiv:2103.01869v1 fatcat:iuofgzxvdfe2dnadv6as6e5744

Generation of high order geometry representations in Octree meshes

Harald G. Klimach, Jens Zudrop, Sabine P. Roller
2015 PeerJ Computer Science  
. • Sabine P. Roller wrote the paper, reviewed drafts of the paper.  ... 
doi:10.7717/peerj-cs.35 fatcat:w3kicwwk7ngg7m7pvh5ugkpmna

Feature Norms of German Noun Compounds

Stephen Roller, Sabine Schulte im Walde
2014 Proceedings of the 10th Workshop on Multiword Expressions (MWE)  
., 2009; Silberer and Lapata, 2012; Roller and Schulte im Walde, 2013) . In this paper, we present a new resource of feature norms for a set of 572 concrete, depictable German nouns.  ...  This version is likely to be more useful for research that is highly affected by sparse data, such as multimodal experiments (Andrews et al., 2009; Silberer and Lapata, 2012; Roller and Schulte im Walde  ... 
doi:10.3115/v1/w14-0818 dblp:conf/mwe/RollerW14 fatcat:mymfk5s5d5hq7awwvmwtjxplcq

Die Deutsche Krebshilfe

Helmut Geiger, Klaus Dörrie, Günther Keding, Sabine vonKleist, Hans-Joachim Möhle, Wolfgang Roller, Gerhard Schmidt, Alfons Titzrath
1991 Oncology Research and Treatment  
Im September 1974 gründete Frau Dr. med. Mildred Scheel die Deutsche Krebshilfe, um die Krebsbekämpfung in der Bundesrepublik auf eine breite und bürgernahe Grundlage zu stellen. In den nunmehr 17 Jahren ihres Bestehens ent-wickelte sich die Deutsche Krebshilfe zur größten Bür-gerinitiative unseres Landes auf diesem Gebiet. Knapp ein-einhalb Jahre später, im Februar 1976, wurde die Deutsche Stiftung für Krebsforschung -heute Dr. Mildred Scheel Stiftung für Krebsforschung -ins Leben gerufen. Zu
more » ... iesem Schritt führte die Erkenntnis, daß effektive Krebsbekämpfung ohne eine weitreichende, langfristige Krebsforschung nicht möglich ist. Allgemeine Richtlinien bei der Vergabe von Mitteln seitens der Deutschen Krebshilfe wie auch der Dr. Mildred Scheel Stiftung für Krebsforschung ist dabei die Förderung von in-novativen Projekten in Kliniken und Forschungsein-richtungen. Von ihnen sollen Impulse für weitere Fortschritte in der Krebsbekämpfung ausgehen. Gefördert werden wis-senschaftliche, diagnostische und therapeutische Maß-nahmen in Deutschland; Internationale Kooperationsprojekte sind unter bestimmten Voraussetzungen möglich. Wichtig ist vor allem, daß ein Bezug zur klinischen An-wendung gegeben ist, damit neue Erkenntnisse unmittelbar dem Erkrankten zugute kommen können. Bei der Entscheidung über Förderungsanträge stehen den Vorständen der Deutschen Krebshilfe und der Dr. Mildred Scheel Stiftung für Krebsforschung ein medizinischer und ein wissenschaftlicher Beirat zur Seite. Diesen Gremien gehören fachkundige Wissenschaftler und Ärzte an. Für die bisher ge-förderten mehr als 640 Vorhaben wurden insgesamt 488 Mio. DM finanziert. Hilfsbereitschaft und soziales Engagement der Burger, die die Deutsche Krebshilfe seit ihrer Gründung unterstützt ha-ben, ließen diese Organisation zu einer der erfolgreichsten privaten Institutionen im Gesundheitswesen werden. Als gemeinnütziger Verein ist sie ausschließlich auf Spenden-gelder angewiesen; staatliche Mittel erhält die Deutsche Krebshilfe nicht. Sie kann unabhängig, unbürokratisch, rasch und flexibel reagieren. Wie groß die Bereitschaft der Bevölkerung ist, sich aktiv am Kampf gegen den Krebs zu be-teiligen, zeigt die Gesamtspendensumme von 538 Mio. DM, die seit Bestehen der Deutschen Krebshilfe eingenommen worden ist. Das Spendenaufkommen hat die Deutsche
doi:10.1159/000217049 fatcat:6x3loxh4xnhrrm42pmaipze6ky

Lattice Boltzmann Simulation of non-Darcy Flow in Porous Media

Manuel Hasert, Jörg Bernsdorf, Sabine Roller
2011 Procedia Computer Science  
Flow through porous media at low Reynolds numbers has been studied in detail with the Lattice Boltzmann Method (LBM) for applications such as groundwater flow, pollution transport or adsorption processes. In contrast to that, medium to high Reynolds number flow through porous media, which occurs in many areas of industrial engineering, has not yet widely been investigated on a microscopic level by detailed numerical simulations. In this paper, we focus on air flow through a porous medium,
more » ... e our far goal entails the simulation of acoustic excitations from the turbulent flow leaving the porous medium. We validate the LBM at Reynolds numbers beyond the limit of Darcy's law, and compare the results of direct numerical simulation with those achieved by applying a Smagorinsky-type large eddy turbulence model. For this, we performed flow simulations through a generic (periodic) porous medium at a variety of resolutions to investigate the effect of LES modelling at lower mesh sizes, where the subgrid scale effects become important.
doi:10.1016/j.procs.2011.04.111 fatcat:kvzemeduu5hcbl2ptclvtbzmta

Lattice Boltzmann scheme for electrolytes by an extended Maxwell-Stefan approach

Jens Zudrop, Sabine Roller, Pietro Asinari
2014 Physical Review E  
A Lattice Boltzmann scheme for electrolytes by an extended Maxwell-Stefan approach Jens Zudrop, * Sabine Roller, † and Pietro Asinari ‡ This paper presents an extended multicomponent Lattice Boltzmann  ... 
doi:10.1103/physreve.89.053310 pmid:25353917 fatcat:22xsyfwjpvecbijalbcog64dwy

Heterogeneous Parallelism of Aero-Acoustic Applications Using PACX-MPI

Harald KLIMACH, Sabine P. ROLLER, Claus-Dieter MUNZ
2009 Interdisciplinary Information Sciences  
This paper outlines the distribution of an aeroacoustic application in a heterogeneous super computing environment. PACX-MPI, which is used for this distribution allows the coupling of different architectures without leaving the MPI context in the application itself. This makes the usage of a heterogeneous infrastructure very convenient from the applications point of view. Integrated simulation of fluid flow with its aeroacoustic is a typical multi-scale task with different numerical
more » ... s in the involved parts. The different requirements can be spatially separated as the noise generating object is generally rather small when compared to the area which is to be computed for the sound propagation. A natural division into a domain of noise generation and a domain of noise propagation arises and the two different computational domains are only loosely coupled with numerical requirements very distinct from each other. We demonstrate how the parallel simulation of a 3D aeroacoustic testcase can benefit of the heterogeneity in the infrastructure by mapping the heterogeneous computational domain on the appropriate architectures.
doi:10.4036/iis.2009.79 fatcat:s2b3ackqwjfcph7mpvq46lxx4m

The Teraflop Workbench — Enhancing High Sustained Performance on Vector Systems

Sabine ROLLER, Michael RESCH, Martin GALLE, Wolfgang BEZ
2009 Interdisciplinary Information Sciences  
This paper provides a comprehensive collection of applications from different fields that have all one thing in common: They are typical representatives of their research field with high need for High Performance Computing resources, and they all show high sustained performance on the HPC vector system NEC-SX8. The paper also describes future needs to address when describing the next generation of applications.
doi:10.4036/iis.2009.45 fatcat:kd5pcjvn5fb4jgyfa36ov64mvm

Synergieeffekte durch Kooperation: Hintergründe, Aufgaben und Potentiale des Projekts FoDaKo

Volker Hess, Thomas von Rekowski, Sabine Roller, Nicole Walger
2019 BIBLIOTHEK Forschung und Praxis  
ZusammenfassungVor dem Hintergrund der zunehmenden Bedeutung von Open-Data-Ansätzen und neuen Anforderungen an das Forschungsdatenmanagement (FDM) vonseiten der Förderinstitutionen, streben Forschende und Institutionen einen Wandel bislang praktizierter Vorgehensweisen beim FDM an. In einer universitären Kooperation von drei Rechenzentren und drei Universitätsbibliotheken zielt das Projekt Forschungsdatenmanagement in Kooperation (FoDaKo)Rekowski und Hess (2017), FoDaKo Forschungsprojekt.
more » ... ab, Synergien nutzbar zu machen und Best Practices für das FDM zu entwickeln und auszutauschen. Der Beitrag gibt einen Einblick in Struktur, Maßnahmen und bislang gesammelte Erfahrungen eines Projekts zur Erforschung des Managements von Forschungsdaten in ihrem Lebenszyklus an den Universitäten Düsseldorf, Siegen und Wuppertal und ermöglicht Dritten damit die Adaption von Ansätzen für ein möglichst kosteneffizientes und nachhaltiges FDM.
doi:10.1515/bfp-2019-2009 fatcat:3eercxdvpzczvn5oqku2wonr3m

Transitional flow in intracranial aneurysms - a space and time refinement study below the Kolmogorov scales using Lattice Boltzmann Method [article]

Kartik Jain, Sabine Roller, Kent-Andre Mardal
2015 arXiv   pre-print
., and Roller, S. End-to-end parallel simulations with apes. In Parallel Computing: Accelerating Computational Science and Engineering (CSE) (Munich, Germany, September 2014), M. Bader, A. Bode, H.  ... 
arXiv:1512.02504v1 fatcat:odhl6bkrunhqjlathyw5fyluqe

ExaFSA: Parallel Fluid-Structure-Acoustic Simulation [chapter]

Florian Lindner, Amin Totounferoush, Miriam Mehl, Benjamin Uekermann, Neda Ebrahimi Pour, Verena Krupp, Sabine Roller, Thorsten Reimann, Dörte C. Sternel, Ryusuke Egawa, Hiroyuki Takizawa, Frédéric Simonis
2020 Lecture Notes in Computational Science and Engineering  
In this paper, we present results of the second phase of the project ExaFSA within the priority program SPP1648-Software for Exascale Computing. Our task was to establish a simulation environment consisting of specialized highly efficient and scalable solvers for the involved physical aspects with a particular focus on the computationally challenging simulation of turbulent flow and propagation of the induced acoustic perturbations. These solvers are then coupled in a modular, robust,
more » ... y efficient and fully parallel way, via the open source coupling library preCICE. Whereas we made a first proof of concept for a three-field simulation (elastic structure, surrounding turbulent acoustic flow in the near-field, and pure acoustic wave propagation in the far-field) in the first phase, we removed several scalability limits in the second phase. In particular, we present new contributions to (a) the initialization of communication between processes of F. Lindner · A. Totounferoush · M. Mehl ( ) the involved independent solvers, (b) optimization of the parallelization of data mapping, (c) solver-specific white-box data mapping providing higher efficiency but less flexibility, (d) portability and scalability of the flow and acoustic solvers FASTEST and Ateles on vector architectures by means of code transformation, (e) physically correct information transfer between near-field acoustic flow and farfield acoustic propagation.
doi:10.1007/978-3-030-47956-5_10 fatcat:v4jlceviunah7adlgtzo5mhknu

Utilization of the Brinkman Penalization to Represent Geometries in a High-Order Discontinuous Galerkin Scheme on Octree Meshes

Nikhil Anand, Neda Ebrahimi Pour, Harald Klimach, Sabine Roller
2019 Symmetry  
We investigate the suitability of the Brinkman penalization method in the context of a high-order discontinuous Galerkin scheme to represent wall boundaries in compressible flow simulations. To evaluate the accuracy of the wall model in the numerical scheme, we use setups with symmetric reflections at the wall. High-order approximations are attractive as they require few degrees of freedom to represent smooth solutions. Low memory requirements are an essential property on modern computing
more » ... s with limited memory bandwidth and capability. The high-order discretization is especially useful to represent long traveling waves, due to their small dissipation and dispersion errors. An application where this is important is the direct simulation of aeroacoustic phenomena arising from the fluid motion around obstacles. A significant problem for high-order methods is the proper definition of wall boundary conditions. The description of surfaces needs to match the discretization scheme. One option to achieve a high-order boundary description is to deform elements at the boundary into curved elements. However, creating such curved elements is delicate and prone to numerical instabilities. Immersed boundaries offer an alternative that does not require a modification of the mesh. The Brinkman penalization is such a scheme that allows us to maintain cubical elements and thereby the utilization of efficient numerical algorithms exploiting symmetry properties of the multi-dimensional basis functions. We explain the Brinkman penalization method and its application in our open-source implementation of the discontinuous Galerkin scheme, Ateles. The core of this presentation is the investigation of various penalization parameters. While we investigate the fundamental properties with one-dimensional setups, a two-dimensional reflection of an acoustic pulse at a cylinder shows how the presented method can accurately represent curved walls and maintains the symmetry of the resulting wave patterns.
doi:10.3390/sym11091126 fatcat:bxb2ltjr45f4lgu7cvwhtzcuje

Dynamic Load Balancing for Unstructured Meshes on Space-Filling Curves

Daniel F. Harlacher, Harald Klimach, Sabine Roller, Christian Siebert, Felix Wolf
2012 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops & PhD Forum  
Load imbalance is an important impediment on the path towards higher degrees of parallelism-especially for engineering codes with their highly unstructured problem domains. In particular, when load conditions change dynamically, efficient mesh partitioning becomes an indispensable ingredient of scalable design. However, popular graph-based methods such as those used by ParMetis require global knowledge, which effectively limits the problem size on distributed-memory machines. On such
more » ... es, space-filling curves (SFCs) offer a memory-efficient alternative and many sophisticated schemes have already been proposed. In this paper, we present a simple strategy based on SFCs that is custom-tailored to the needs of static meshes with dynamically changing computational load. Exploiting the properties of this class of problems, it is not only easy to implement but also reduces memory requirements substantially. Moreover, exclusively relying on MPI collective operations, our load-balancing scheme also offers portable performance across a broad range of HPC systems. Experimental evaluation shows excellent scaling behavior for up to 16,384 cores on a Nehalem-Infiniband system and up to 294,912 processes on a Blue Gene/P system.
doi:10.1109/ipdpsw.2012.207 dblp:conf/ipps/HarlacherKRSW12 fatcat:cdi6lz5ygvhjzowt5szfkc5mja

Calculation of low Mach number acoustics: a comparison of MPV, EIF and linearized Euler equations

Sabine Roller, Thomas Schwartzkopff, Roland Fortenbach, Michael Dumbser, Claus-Dieter Munz
2005 Mathematical Modelling and Numerical Analysis  
The calculation of sound generation and propagation in low Mach number flows requires serious reflections on the characteristics of the underlying equations. Although the compressible Euler/Navier-Stokes equations cover all effects, an approximation via standard compressible solvers does not have the ability to represent acoustic waves correctly. Therefore, different methods have been developed to deal with the problem. In this paper, three of them are considered and compared to each other.
more » ... are the Multiple Pressure Variables Approach (MPV), the Expansion about Incompressible Flow (EIF) and a coupling method via heterogeneous domain decomposition. In the latter approach, the non-linear Euler equations are used in a domain as small as possible to cover the sound generation, and the locally linearized Euler equations approximated with a high-order scheme are used in a second domain to deal with the sound propagation. Comparisons will be given in construction principles as well as implementational effort and computational costs on actual numerical examples. Mathematics Subject Classification. 41A60, 65M55, 76Q05.
doi:10.1051/m2an:2005016 fatcat:wrcpnzm4n5gwnhhghgv4tbb6ry

Compressible flow simulation with moving geometries using the Brinkman penalization in high-order Discontinuous Galerkin

Neda Ebrahimi Pour, Nikhil Anand, Harald Klimach, Sabine Roller
2021 Advanced Modeling and Simulation in Engineering Sciences  
AbstractIn this work we investigate the Brinkman volume penalization technique in the context of a high-order Discontinous Galerkin method to model moving wall boundaries for compressible fluid flow simulations. High-order approximations are especially of interest as they require few degrees of freedom to represent smooth solutions accurately. This reduced memory consumption is attractive on modern computing systems where the memory bandwidth is a limiting factor. Due to their low dissipation
more » ... d dispersion they are also of particular interest for aeroacoustic problems. However, a major problem for the high-order discretization is the appropriate representation of wall geometries. In this work we look at the Brinkman penalization technique, which addresses this problem and allows the representation of geometries without modifying the computational mesh. The geometry is modelled as an artificial porous medium and embedded in the equations. As the mesh is independent of the geometry with this method, it is not only well suited for high-order discretizations but also for problems where the obstacles are moving. We look into the deployment of this strategy by briefly discussing the Brinkman penalization technique and its application in our solver and investigate its behavior in fundamental one-dimensional setups, such as shock reflection at a moving wall and the formation of a shock in front of a piston. This is followed by the application to setups with two and three dimensions, illustrating the method in the presence of curved surfaces.
doi:10.1186/s40323-021-00195-4 fatcat:sflnplx3wbhmfkk426x64qziuy
« Previous Showing results 1 — 15 out of 1,758 results