In this Doktorarbeit the Lattice Boltzmann scheme, a heuristic method for the simulation of flows in complicated boundaries, is investigated. Its theory is renewed by emphasizing the entropy maximization principle, and new means for the modelling of geometries (including moving boundaries) and the visual representation of evoluting flows are presented. An object oriented implemen- tation is given with communication between objects realized by an interpreter object and communication from outside realized via interprocess communica- tion. Within the new theoretical apprach the applicability of existing Lattice Boltzmann schemes to model thermal flows for arbitrary temperatures is reex- amined. / In dieser Doktorarbeit wird das Gitter-Boltzmann-Schema, eine heuristische Methode fuer die Simulation von Stroemungen innerhalb komplexer Raender, untersucht. Die zugrundeliegende Theorie wird unter neuen Gesichtspunkten, insbesondere dem Prinzip der Entropiemaximierung, betrachtet. Des weiteren werden neuartige Methoden fuer die Modellierung der Geometrie (einschl. beweglicher Raender) und der visuellen Darstellung aufgezeigt. Eine objektorientierte Implementierung wird vorgestellt, wobei die Kommunikation zwischen den Objekten über Interpreter-Objekte und die Kommunikation mit der Aussenwelt ueber Interprozess-Kommunikation gehandhabt wird. Mit dem neuen theoretischen Ansatz wird die Gueltigkeit bestehender Gitter-Boltzmann-Schemata fuer die Anwendung auf Stroemungen mit nicht konstanter Temperatur untersucht.

Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-01-11T14:54:52Z No. of bitstreams: 1 andresricardovaldez.pdf: 6592036 bytes, checksum: 23a86a3d84f13bffa421f219e7e4501d (MD5) / Rejected by Fabíola Rubim (fabiola.rubim@ufjf.edu.br), reason: on 2018-01-12T11:05:10Z (GMT) / Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-01-12T11:46:32Z No. of bitstreams: 1 andresricardovaldez.pdf: 6592036 bytes, checksum: 23a86a3d84f13bffa421f219e7e4501d (MD5) / Rejected by Adriana Oliveira (adriana.oliveira@ufjf.edu.br), reason: Favor corrigir: Membro da banca: Filho, José Karam on 2018-01-23T14:01:35Z (GMT) / Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-01-23T14:06:58Z No. of bitstreams: 1 andresricardovaldez.pdf: 6592036 bytes, checksum: 23a86a3d84f13bffa421f219e7e4501d (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2018-01-23T14:22:22Z (GMT) No. of bitstreams: 1 andresricardovaldez.pdf: 6592036 bytes, checksum: 23a86a3d84f13bffa421f219e7e4501d (MD5) / Made available in DSpace on 2018-01-23T14:22:22Z (GMT). No. of bitstreams: 1 andresricardovaldez.pdf: 6592036 bytes, checksum: 23a86a3d84f13bffa421f219e7e4501d (MD5) Previous issue date: 2017-09-18 / Neste trabalho são apresentados aspectos de modelagem computacional para o estudo de Interação Fluido-Estrutura (FSI). Numericamente, o Método de Lattice Boltzmann (LBM) é usado para resolver a mecânica dos fluidos, em particular as equações de Navier-Stokes incompressíveis. Neste contexto, são abordados problemas de escoamentos complexos, caracterizado pela presença de obstáculos. A imposição das restrições na interface fluido-sólido é feita utilizando princípios variacionais, empregando o Princípio de Balanço de Potências Virtuais (PVPB) para obter as equações de Euler-Lagrange. Esta metodologia permite determinar as dependências entre carregamentos cinematicamente compatíveis e o estado mecânico adotado. Neste sentido, as condições de interface fluido-sólido são abordadas pelo Método de Fronteira Imersa (IBM) visando técnicas computacionais de baixo custo. A metodologia IBM trata o equilíbrio das equações na interface fluido-sólido através da interpolação entre os nós Lagrangianos (sólidos) e os nós Eulerianos (fluidos). Neste contexto, uma modificação desta estratégia que fornece soluções mais precisas é estudada. Para mostrar as capacidades do acoplamento LBM-IBM são apresentados vários experimentos computacionais que demonstram grande fidelidade entre as soluções obtidas e as soluções disponíveis na literatura. / This work presents computational modeling aspects for studying Fluid-Structure Interaction (FSI). The Lattice Boltzmann Method (LBM) is employed to solve the fluid mechanics considering the incompressible Navier-Stokes equations. The flows studied are complex due to the presence of arbitrary shaped obstacles. The obstacles alters the bulk flow adding complexity to the analysis. In this work the Euler-Lagrange equations are obtained employing the Principle of Virtual Power Balance (PVPB). Consequently, the functional dependencies between the mechanical state and every kinematic compatible loadings are established employing variational arguments. This modeling technique allows to study the fluid-solid boundary constraint. In this context the fluid-solid interface is handled employing the Immersed Boundary Method (IBM). The IBM deals with the fluid-solid interface equilibrium equations performing an interpolation of forces between Lagrangian nodes (solid domain) and Eulerian Lattice grid (fluid domain). In this work a different version of this methodology is studied that allows to obtain more accurate solutions. To show the capabilities of the implemented LBM-IBM solver several experiments are done showing the agreement with the benchmarks results available in literature.

Lattice-Boltzmann method(LBM), a promising new particle-based simulation technique for complex and multiscale fluid flows, has seen tremendous adoption in recent years in computational fluid dynamics. Even with a state-of-the-art LBM solver such as Palabos, a user still has to manually write his program using the library-supplied primitives. We propose an automated code generator for a class of LBM computations with the objective to achieve high performance on modern architectures. Tiling is a very important loop transformation used to improve the performance of stencil computations by exploiting locality and parallelism. In the first part of the work, we explore diamond tiling, a new tiling technique to exploit the inherent ability of most stencils to allow tile-wise concurrent start. This enables perfect load-balance during execution and reduces the frequency of synchronization required. Few studies have looked at time tiling for LBM codes. We exploit a key similarity between stencils and LBM to enable polyhedral optimizations and in turn time tiling for LBM. Besides polyhedral transformations, we also describe a number of other complementary transformations and post processing necessary to obtain good parallel and SIMD performance on modern architectures. We also characterize the performance of LBM with the Roofline performance model. Experimental results for standard LBM simulations like Lid Driven Cavity, Flow Past Cylinder, and Poiseuille Flow show that our scheme consistently outperforms Palabos–on average by3 x while running on 16 cores of a n Intel Xeon Sandy bridge system. We also obtain a very significant improvement of 2.47 x over the native production compiler on the SPECLBM benchmark.

Dans les structures en béton précontraint, les phénomènes de fluage et de retrait tendent à réduire les efforts de précontrainte initialement prévus pour maintenir le béton dans un état minimisant les forces de traction et donc la fissuration. La compréhension et la prédiction de ces phénomènes par le biais de modèles sont donc primordiales pour la conception et la maintenance à long terme des ouvrages du génie civil tels que les enceintes de confinement des centrales nucléaires.L’objectif de cette thèse est d’élaborer un cadre de modélisation micromécanique pour décrire de manière unifiée le retrait et le fluage dans les matériaux cimentaires. Pour cela, l’étude se base sur l’échelle de la microstructure poreuse du gel de C-S-H où les mécanismes intrinsèques de ces déformations différées du béton opèrent. Une approche d’homogénéisation numérique modélisant ces phénomènes dans des microstructures poreuses à morphologies quelconques est développée. Une description explicite du réseau poreux ainsi que de la phase liquide de l’eau pendant les processus de séchage/humidification est prise en compte. Les mécanismes concernant lesdéformations différées dans la phase solide sont modélisés par la théorie de la microprécontrainte-solidification (MPS). Les simulations à l’échelle microscopique sont réalisées par une approche originale couplant la méthode de Lattice Boltzmann (LBM) et la méthode des éléments finis (FEM). La LBM est utilisée pour décrire la distribution du liquide capillaire à l’échelle du pore,tandis que la FEM est employée pour simuler la déformation du squelette solide sous l’action combinée de l’eau dans l’espace poreux et d’un chargement macroscopique.La démarche proposée permet, au travers des simulations, de mieux comprendre les mécanismes liés à la non saturation et aux effets capillaires dans les milieux poreux. En particulier, la prise en compte de morphologies réalistes de microstructures et des ménisques formés conduit à différents régimes de retrait/gonflement. Ainsi les effets de l’intensité de la pression capillaire,de la tension de surface et des surfaces de chargement sur la réponse élastique du squelette solide sont évalués. Enfin, nous proposons une extension des approches précédentes au cas d’un squelette viscoélastique se déformant sous les effets de la pression capillaire et des tensions de surface. A partir des observations numériques réalisées, nous proposons un modèle pour décrire le fluage et le retrait du gel de C-S-H de manière unifiée / In pre-stressed concrete structures, creep and shrinkage tend to reduce the pre-stress forces which are initially produced so as to maintain concrete in a state minimizing traction forces and then cracks. Understanding and predicting these phenomena through models are thus highly important for the design and durability of civil engineering structures, such as containment buildings in nuclear power plants.The objective of this thesis is to develop a micromechanical modeling framework to describe shrinkage and creep in cementitious materials in a unified manner. For this purpose, the study focuses on the scale of the porous structure of the C-S-H gel where the intrinsic mechanisms of delayed strains are active. A computational homogenization approach is developed to model these phenomena in porous structures with arbitrary morphologies. An explicit description of the porous network and of the liquid phase of water during the drying/humidification process is taken into account. The mechanisms related to delayed strains in the solid phase are modeled by the microprestress-solidification theory (MPS). The simulations at the microscale are conductedbased on an original approach coupling the Lattice Boltzmann method (LBM) and the finite element method (FEM). The LBM is used to describe the distribution of capillary water in the porous structure, whereas the FEM serves as modeling the strain of the solid skeleton under the capillary water effets and a macroscopic load.The proposed method allows, by means of the simulations, to better understand the mechanisms related to the capillary effects in the porous structure. More specifically, taking into account realistic morphologies of microstructures and of the formed menisci lead to different regimes of shrinkage/swelling. Then, the effects of capillary pressure intensity, of surface tension and of morphologies of capillary surfaces on the elastic response of the solid skeleton are evaluated. Finally, the above approaches are extended to the case of a viscoelastic solid deformed under the action of the capillary water. From numerical observations, we propose a model is proposed to describe the creep and shrinkage of C-S-H gel in a unified way

L’évolution d’une phase microbienne au sein d’un milieu poreux est un processus complexe de par la prise en compte des effets de croissance (ou de mortalité) et d’étalement de la phase cellulaire. D’autres processus tels que l’arrachement d’une partie du biofilm ou l’attachement-détachement de cellules mobiles depuis la phase fluide peuvent aussi contribuer à la variation du volume de biofilm présent. Une meilleure compréhension des interactions mis en jeu entre les processus de croissance de biofilm, du transport de soluté et de l’écoulement et une modélisation rigoureuse de ce processus de croissance à l’échelle microscopique est un enjeu essentiel à une prédiction plus fine du devenir des polluants dans les sols. L’évolution temporelle d’un milieu poreux sous l’effet de l’activité biologique constitue toutefois à l’heure actuelle un défi scientifique majeur d’un point de vue de la modélisation numérique. Les variations locales de la géométrie du domaine (bio-obstruction des pores) induisent en effet une chenalisation de l’écoulement et du transport qui va évoluer au cours du temps. Si différentes méthodes numériques – lagrangiennes ou eulériennes – ont été développées (méthode de capture du front, méthode d’interface diffuse de type « Level Set » ou « Volume Of Fluid »), elles restent souvent peu adaptées à des modélisations 3D à l’échelle du pore (temps de calcul, remaillage parfois nécessaire, problème de gain ou de perte de masse). Nous combinons ici une méthode IBM (Immersed Boundary Method) à une méthode LBM (Lattice Boltzman Method) pour le calcul de l’écoulement en 3D tandis qu’une approche de type VOF (Volume of Fluid) ou par reconstruction d’interface couplée à une discrétisation en Volume Finis est utilisée pour le transport des espèces chimiques. L’intérêt ici de la méthode IB-LBM est de pouvoir bénéficier de la précision de la formulation Lattice- Boltzmann tout en travaillant sur un maillage fixe, un terme correcteur venant modifier la vitesse au voisinage des interfaces mobiles. Le modèle d’écoulement-transport en milieu poreux évolutif développé est ensuite couplé à un modèle d’automate cellulaire prenant en compte les processus d’attachement-détachement. Le modèle est comparé à des benchmarks numériques et utilisé pour étudier les différents régimes de croissance du biofilm en fonction des conditions hydrodynamiques. Dans le dernier chapitre, ce modèle est étendu à la prise en compte d’une phase non-miscible afin d’étudier l’impact des processus de biodégradation sur la dissolution d’une phase polluante piégé. On se limite aux conditions où le NAPL est à saturation résiduelle. L’influence de la production de biosurfactant sur la solubilité du polluant ainsi que la toxicité de celui-ci sur la cinétique de croissance des bactéries est prise en compte. Plusieurs résultats numériques sont présentés afin d’illustrer l’influence des différents paramètres hydrodynamiques sur la dissolution du NAPL. / Mathematical modeling of transport in porous media of organic chemical species in the presence of a bacterial population growing in the form of biofilms is an important area of research for environmental and industrial applications such as the treatment and the remediation of groundwater contaminated by organic pollutants. Biofilms, which are composed of bacteria and extracellular organic substances, grow on the pore-walls of the porous medium. Biodegradable organic solutes are converted into biomass or other organic compounds by the bacterial metabolism. This evolution of the microbial biomass phase within the porous medium is a complex process due mainly to growth (or decay) and spatial spreading of the cellular phase. Processes such as biofilm sloughing and attachment (or detachment) of cells from the fluid phase may also contribute to the biofilm volume variation. In this context, the aim of the thesis is to focus on the mechanisms that control the development of biofilms in porous media and its impact on the hydrodynamic properties of the porous matrix. The objective of this work is to model this pore-scale phenomenon of biofilm growth by integrating the various mechanisms which favor the bacterial development (bacterial proliferation, assimilation of nutrients to synthesize new cellular materials, attachment of cells) or, conversely, which are responsible for slowing down (e.g., detachment of cells, toxicity). An IB-LB model is developed for flow calculation and non-boundary conforming finite volume methods (volume of fluid and reconstruction methods) are used for reactive solute transport. A sophisticated cellular automaton model is developed to describe the spatial distribution of bacteria. Several numerical simulations have been performed on complex porous media and a quantitative diagram representing the transitions between the different biofilm growth patterns was proposed. Finally, the bioenhanced dissolution of NAPL in the presence of biofilms was simulated at the pore scale. The impact of biosurfactants and NAPL toxicity on bacterial growth has been investigated.

Ce travail de thèse a pour objet l’étude numérique et expérimentale de la déstabilisation de milieux granulaires immergés par fluidisation. Cette instabilité hydromécanique est un mécanisme précurseur de l’érosion régressive, processus de dégradation au coeur de la problématique de l’érosion interne des ouvrages hydrauliques en terre. La compréhension de ces mécanismes d’érosion nécessite une description rigoureuse du couplage et de l’interaction entre le fluide et les particules de sol. A cette fin, un modèle 2D a été utilisé en couplant deux méthodes particulaires, la méthode des éléments discrets (DEM) pour modéliser le comportement mécanique de la phase solide et la méthode Lattice Boltzmann (LBM) pour la phase fluide. Des expériences servant de validation à cette simulation numérique 2D ont également été réalisées en s’appuyant sur une technique de visualisation interne d’un empilement granulaire combinant l’ajustement d’indice de réfraction des deux phases et la fluorescence induite par plan laser. / The subject of this thesis is the numerical analysis and experimental investigation of the destabilization of submerged granular media caused by fluidization. This hydromechanical instability is one of the mechanisms that may trigger the regressive erosion, which is one of the main degradation phenomena driving the internal erosion of earthen hydraulic constructions. Such erosion mechanisms can only be understood through a rigorous description of the coupling and interaction between the eroding fluid and the soil particles. For this purpose, a 2D model has been used coupling two different numerical techniques, namely the discrete element method (DEM) for modelling the mechanical behaviour of the solid phase and the Lattice Boltzmann method (LBM) for the fluid phase. The experimental validation of this numerical 2D simulation has been carried out using two optical techniques for the internal visualization of a granular sample, namely the adjustment of the refraction index of the two phases and the laser-induced fluorescence.

Cette thèse porte sur le développement d’une nouvelle technique de modélisation des problèmes IFS utilisant les méthodes particulaires. Ce travail s’inscrit dans la continuité des travaux de recherche de l’équipe biomécanique du LAMIH, concernant la compréhension du comportement de l’os humain dans son environnement de moelle osseuse. La méthode SPH a été utilisée pour la modélisation des travées osseuses, supposées dans une première approche comme des solides élastiques. La méthode LB a été développée pour la modélisation des écoulements de moelle considérée comme un fluide visqueux incompressible. L’efficacité et la performance de ces deux méthodes ont été démontrées grâce aux benchmarks académiques évalués et les résultats comparés à ceux de la littérature ou ceux obtenus par des logiciels commerciaux. A l’issue d’une revue de l’état de l’art des techniques de couplage fluide-structure, une approche partitionnée en temps a été choisie, permettant d’utiliser deux codes distincts basés sur des algorithmes de résolution de type dynamique explicite. La discrétisation spatiale est faite par une technique spécifique basée sur les domaines fictifs, cette technique est très efficace car elle ne nécessite pas de rediscrétisation des domaines. L’approche de couplage développée a été appliquée à des benchmarks académiques ainsi qu’à une application en biomécanique, ayant permis d’aboutir à des résultats numériques satisfaisants. Plusieurs pistes d’amélioration sont maintenant nécessaires afin d’aller vers des modélisations plus biofidèles telles que la prise en compte du contact et de l’endommagement. / The objective of this thesis is the development of a new technique for the FSI problems modelling using particulars methods. This work is in the continuity of the LAMIH biomechanics team research works, regarding the comprehension of behavior of bone in its environment of marrow. The SPH method was used for the trabeculae modelling, supposed in a first attempt as an elastic solid. The LB method was developed for the marrow flow modelling considered as a viscous incompressible liquid. The efficacy and performance of these two methods were demonstrated using academics benchmarks which were evaluated and the results were compared of those of literature and of those obtained from commercials softwares. Following a bibliographic review of FSI coupling techniques, a partitioned approach in time was chosen, allowing the use of two separates codes, both based on a dynamic explicit algorithm resolution scheme. The special discretization was done based on a specific technique of fictional domain, this technique is very efficient because it doesn’t require an additional domain discretization. The coupling approach developed was applied on academic benchmarks and on a biomechanical application, leading to satisfactory numerical results. Many Improvement track are now necessary to go towards more biofidelic modeling as taking into account the contact and the damage.

In this Doktorarbeit the Lattice Boltzmann scheme, a heuristic method for the simulation of flows in complicated boundaries, is investigated. Its theory is renewed by emphasizing the entropy maximization principle, and new means for the modelling of geometries (including moving boundaries) and the visual representation of evoluting flows are presented. An object oriented implemen- tation is given with communication between objects realized by an interpreter object and communication from outside realized via interprocess communica- tion. Within the new theoretical apprach the applicability of existing Lattice Boltzmann schemes to model thermal flows for arbitrary temperatures is reex- amined. / In dieser Doktorarbeit wird das Gitter-Boltzmann-Schema, eine heuristische Methode fuer die Simulation von Stroemungen innerhalb komplexer Raender, untersucht. Die zugrundeliegende Theorie wird unter neuen Gesichtspunkten, insbesondere dem Prinzip der Entropiemaximierung, betrachtet. Des weiteren werden neuartige Methoden fuer die Modellierung der Geometrie (einschl. beweglicher Raender) und der visuellen Darstellung aufgezeigt. Eine objektorientierte Implementierung wird vorgestellt, wobei die Kommunikation zwischen den Objekten über Interpreter-Objekte und die Kommunikation mit der Aussenwelt ueber Interprozess-Kommunikation gehandhabt wird. Mit dem neuen theoretischen Ansatz wird die Gueltigkeit bestehender Gitter-Boltzmann-Schemata fuer die Anwendung auf Stroemungen mit nicht konstanter Temperatur untersucht.

Cancer cells display characteristic traits acquired in a step-wise manner during carcinogenesis. Some of these traits are autonomous growth, induction of angiogenesis, invasion and metastasis. In this thesis, the focus is on one of the latest stages of tumor progression, tumor invasion. Tumor invasion emerges from the combined effect of tumor cell-cell and cell-microenvironment interactions, which can be studied with the help of mathematical analysis. Cellular automata (CA) can be viewed as simple models of self-organizing complex systems in which collective behavior can emerge out of an ensemble of many interacting "simple" components. In particular, we focus on an important class of CA, the so-called lattice-gas cellular automata (LGCA). In contrast to traditional CA, LGCA provide a straightforward and intuitive implementation of particle transport and interactions. Additionally, the structure of LGCA facilitates the mathematical analysis of their behavior. Here, the principal tools of mathematical analysis of LGCA are the mean-field approximation and the corresponding Lattice Boltzmann equation. The main objective of this thesis is to investigate important aspects of tumor invasion, under the microscope of mathematical modeling and analysis: Impact of the tumor environment: We introduce a LGCA as a microscopic model of tumor cell migration together with a mathematical description of different tumor environments. We study the impact of the various tumor environments (such as extracellular matrix) on tumor cell migration by estimating the tumor cell dispersion speed for a given environment. Effect of tumor cell proliferation and migration: We study the effect of tumor cell proliferation and migration on the tumor’s invasive behavior by developing a simplified LGCA model of tumor growth. In particular, we derive the corresponding macroscopic dynamics and we calculate the tumor’s invasion speed in terms of tumor cell proliferation and migration rates. Moreover, we calculate the width of the invasive zone, where the majority of mitotic activity is concentrated, and it is found to be proportional to the invasion speed. Mechanisms of tumor invasion emergence: We investigate the mechanisms for the emergence of tumor invasion in the course of cancer progression. We conclude that the response of a microscopic intracellular mechanism (migration/proliferation dichotomy) to oxygen shortage, i.e. hypoxia, maybe responsible for the transition from a benign (proliferative) to a malignant (invasive) tumor. Computing in vivo tumor invasion: Finally, we propose an evolutionary algorithm that estimates the parameters of a tumor growth LGCA model based on time-series of patient medical data (in particular Magnetic Resonance and Diffusion Tensor Imaging data). These parameters may allow to reproduce clinically relevant tumor growth scenarios for a specific patient, providing a prediction of the tumor growth at a later time stage. / Krebszellen zeigen charakteristische Merkmale, die sie in einem schrittweisen Vorgang während der Karzinogenese erworben haben. Einige dieser Merkmale sind autonomes Wachstum, die Induktion von Angiogenese, Invasion und Metastasis. Der Schwerpunkt dieser Arbeit liegt auf der Tumorinvasion, einer der letzten Phasen der Tumorprogression. Die Tumorinvasion ensteht aus der kombinierten Wirkung von den Wechselwirkungen Tumorzelle-Zelle und Zelle-Mikroumgebung, die mit die Hilfe von mathematischer Analyse untersucht werden können. Zelluläre Automaten (CA) können als einfache Modelle von selbst-organisierenden komplexen Systemen betrachtet werden, in denen kollektives Verhalten aus einer Kombination von vielen interagierenden "einfachen" Komponenten entstehen kann. Insbesondere konzentrieren wir uns auf eine wichtige CA-Klasse, die sogenannten Zelluläre Gitter-Gas Automaten (LGCA). Im Gegensatz zu traditionellen CA bieten LGCA eine einfache und intuitive Umsetzung der Teilchen und Wechselwirkungen. Zusätzlich erleichtert die Struktur der LGCA die mathematische Analyse ihres Verhaltens. Die wichtigsten Werkzeuge der mathematischen Analyse der LGCA sind hier die Mean-field Approximation und die entsprechende Lattice - Boltzmann - Gleichung. Das wichtigste Ziel dieser Arbeit ist es, wichtige Aspekte der Tumorinvasion unter dem Mikroskop der mathematischen Modellierung und Analyse zu erforschen: Auswirkungen der Tumorumgebung: Wir stellen einen LGCA als mikroskopisches Modell der Tumorzellen-Migration in Verbindung mit einer mathematischen Beschreibung der verschiedenen Tumorumgebungen vor. Wir untersuchen die Auswirkungen der verschiedenen Tumorumgebungen (z. B. extrazellulären Matrix) auf die Migration von Tumorzellen dürch Schätzung der Tumorzellen-Dispersionsgeschwindigkeit in einem gegebenen Umfeld. Wirkung von Tumor-Zellenproliferation und Migration: Wir untersuchen die Wirkung von Tumorzellenproliferation und Migration auf das invasive Verhalten der Tumorzellen durch die Entwicklung eines vereinfachten LGCA Tumorwachstumsmodells. Wir leiten die entsprechende makroskopische Dynamik und berechnen die Tumorinvasionsgeschwindigkeit im Hinblick auf die Tumorzellenproliferation- und Migrationswerte. Darüber hinaus berechnen wir die Breite der invasiven Zone, wo die Mehrheit der mitotischer Aktivität konzentriert ist, und es wird festgestellt, dass diese proportional zu den Invasionsgeschwindigkeit ist. Mechanismen der Tumorinvasion Entstehung: Wir untersuchen Mechanismen, die für die Entstehung von Tumorinvasion im Verlauf des Krebs zuständig sind. Wir kommen zu dem Schluss, dass die Reaktion eines mikroskopischen intrazellulären Mechanismus (Migration/Proliferation Dichotomie) zu Sauerstoffmangel, d.h. Hypoxie, möglicheweise für den Übergang von einem gutartigen (proliferative) zu einer bösartigen (invasive) Tumor verantwortlich ist. Berechnung der in-vivo Tumorinvasion: Schließlich schlagen wir einen evolutionären Algorithmus vor, der die Parameter eines LGCA Modells von Tumorwachstum auf der Grundlage von medizinischen Daten des Patienten für mehrere Zeitpunkte (insbesondere die Magnet-Resonanz-und Diffusion Tensor Imaging Daten) ermöglicht. Diese Parameter erlauben Szenarien für einen klinisch relevanten Tumorwachstum für einen bestimmten Patienten zu reproduzieren, die eine Vorhersage des Tumorwachstums zu einem späteren Zeitpunkt möglich machen.

<div>As high performance computing systems are advancing from petascale to exascale, scientific workflows to integrate simulation and visualization/analysis are a key factor to influence scientific campaigns. As one of the campaigns to study fluid behaviors, computational fluid dynamics (CFD) simulations have progressed rapidly in the past several decades, and revolutionized our lives in many fields. Lattice Boltzmann method (LBM) is an evolving CFD approach to significantly reducing the complexity of the conventional CFD methods, and can simulate complex fluid flow phenomena with cheaper computational cost. This research focuses on accelerating the workflow of LBM simulation and data analysis.</div><div><br></div><div>I start my research on how to effectively integrate each component of a workflow at extreme scales. Firstly, we design an in-situ workflow benchmark that integrates seven state-of-the-art in-situ workflow systems with three synthetic applications, two real-world CFD applications, and corresponding data analysis. Then detailed performance analysis using visualized tracing shows that even the fastest existing workflow system still has 42% overhead. Then, I develop a novel minimized end-to-end workflow system, Zipper, which combines the fine-grain task parallelism of full asynchrony and pipelining. Meanwhile, I design a novel concurrent data transfer optimization method, which employs a multi-threaded work-stealing algorithm to transfer data using both channels of network and parallel file system. It significantly reduces the data transfer time by up to 32%, especially when the simulation application is stalled. Then investigation on the speedup using OmniPath network tools shows that the network congestion has been alleviated by up to 80%. At last, the scalability of the Zipper system has been verified by a performance model and various largescale workflow experiments on two HPC systems using up to 13,056 cores. Zipper is the fastest workflow system and outperforms the second-fastest by up to 2.2 times.</div><div><br></div><div>After minimizing the end-to-end time of the LBM workflow, I began to accelerate the memory-bound LBM algorithms. We first design novel parallel 2D memory-aware LBM algorithms. Then I extend to design 3D memory-aware LBM that combine features of single-copy distribution, single sweep, swap algorithm, prism traversal, and merging multiple temporal time steps. Strong scalability experiments on three HPC systems show that 2D and 3D memory-aware LBM algorithms outperform the existing fastest LBM by up to 4 times and 1.9 times, respectively. The speedup reasons are illustrated by theoretical algorithm analysis. Experimental roofline charts on modern CPU architectures show that memory-aware LBM algorithms can improve the arithmetic intensity (AI) of the fastest existing LBM by up to 4.6 times.</div>