Neste trabalho foi avaliada a utilização da grid computing em aspectos importantes para simulações em Física Computacional. Em particular, para aplicações de diagonalização de matrizes de grande porte. O projeto de código aberto Globus Toolkit foi utilizado para comparar o desempenho da biblioteca paralela de álgebra linear ScaLAPACK em duas versões baseadas na biblioteca de passagem de mensagens, a versão tradicional MPICH e a versão desenvolvida para um ambiente de grid computing MPICH-G2. Várias simulações com diagonalização de matrizes complexas de diversos tamanhos foram realizadas. Para um sistema com uma matriz de tamanho 8000 x 8000 distribuída em 8 processos, nos nós de 64 bits foi alcançado um speedup de 7,71 com o MPICH-G2. Este speedup é muito próximo do ideal que, neste caso, seria igual a 8. Foi constatado também que a arquitetura de 64 bits tem melhor desempenho que a de 32 bits nas simulações executadas para este tipo de aplicação / This work evaluates the use of grid computing in essential issues related to Computational Physics simulations. In particular, for applications with large scale matrix diagonalization. The Globus Toolkit open source project was used to compare the performance of the linear algebra parallel library ScaLAPACK in two different versions based on the message passing library, the traditional version MPICH and its version developed for a grid computing environment MPICH-G2. Several simulations within large scale diagonalization of complex matrix were performed. A 7.71 speedup was reached with the MPICH-G2 for a 8000 x 8000 size matrix distributed in 8 processes on 64 bits nodes. This was very close to the ideal speedup, that would be in this case, 8. It was also evidenced that the 64 bits architecture has better performance than the 32 bits on the performed simulations for this kind of application.

Nous avons étudié les récepteurs GABAA dans un modèle de la maladie de Huntington. En combinant des approches biochimiques, moléculaires, électrophysiologiques et de l’imagerie haute résolution, nous avons montré une modification de la neurotransmission GABAergique chez des animaux à des stades pre- et post-symptomatiques. Nos études montrent une diminution de de la neurotransmission GABAergique dans le globus pallidus des souris Huntington qui pourrait conduire à une modification des noyaux de sortie des ganglions de la base et de l’activité motrice. L’ensemble de nos résultats permet de définir le rôle de différents types de récepteurs GABAA dans le cerveau dans des conditions physiologiques et pathologiques. / We explored GABAergic neurotransmission in a mouse model of Huntington's disease. Combining molecular, imaging and electrophysiologicaltechniques, we showed changes of GABAergic neurotransmission in presymptomatic and symptomatic R6/1 mice. Our data demonstrated a decreased GABAergic inhibition in the globus pallidus of R6/1 mice, which could result in an alteration of basal ganglia output nuclei and motor activity. Taken together, our results will help to define the contribution of receptor subtypes to inhibitory transmission throughout the brain in physiological and pathophysiological states.

Chan, Chi Yung Savio. / "January 2002." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2002. / Includes bibliographical references (p. 130-182). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

La utilización de aplicaciones de diferente naturaleza dentro de un mismo entorno, entorno heterogéneo, se está extendiendo gracias a la incorporación de técnicas de virtualización a los servidores. Compartir un servidor ofrece ventajas sobretodo en términos de eficiencia de energía, utilización del espacio o mantenimiento. La virtualización añade ventajas en la separación de las diferentes aplicaciones o entornos. Aún así los gestores de recursos para entornos heterogéneos tienen como principal dificultad ofrecer calidad de servicio (QoS) a diferentes aplicaciones, entornos o cargas. Una aplicación que realice streaming y otra que realice cálculo intensivo, normalmente , no colisionaran ya que los recursos utilizados son diferentes. Por el otro lado, colisionaran dos aplicaciones que trabajen con la CPU.Nuestra propuesta ofrece la posibilidad de introducir dentro de estos gestores de recursos la capacidad de predecir este tipo de entornos, en concreto transaccionales y Grid, para aumentar la QoS y el rendimiento. Las predicciones han de utilizar técnicas de simulación ya que la mayoria de las veces el sistema no será representable mediante técnicas analíticas, por ser un sistema saturado o tener características difíciles de representar.La simulación es una técnica utilizada para predecir el comportamiento de sistemas en multitud de áreas. Las simulaciones de componentes hardware son muy comunes, dado el coste de construcción de los sistemas simulados (procesadores, memorias...). Sin embargo, el uso de la simulación en entornos complejos, como es el middleware, y su aplicación en gestores de recursos tiene un uso muy bajo. Nosotros proponemos simulaciones ligeras capaces de obtener resultados utilizables en estos entornos.Entre las aportaciones y contribuciones de la tesis tenemos: (i) utilización de métodos de simulación para incrementar el rendimiento y la calidad de servicio de estos sistemas. (ii) ampliación de un sistema de monitorización global para aplicaciones mixtas (JAVA y C) que nos ofrece la posibilidad de conseguir información de lo que ocurre en el middleware y de relacionarlo con el sistema. (iii) creación de un gestor de recursos capaz de repartir los recursos en un entorno heterogéneo utilizando la predicción para tener en cuenta diferentes parámetros de calidad de servicio.En la tesis se muestran los mecanismos de creación de los distintos simuladores, las herramientas de obtención de datos y monitorización, así como mecanismos autónomos que pueden alimentarse de la predicción para producir mejores resultados. Los resultados obtenidos, con gran impacto en la QoS en el gestor creado para Globus, demuestran que los métodos aplicados en esta tesis pueden ser válidos para crear gestores de recursos inteligentes, alimentados de las predicciones del sistema para tomar decisiones. Finalmente, utilizamos las simulaciones realizadas incorporándolas dentro de un prototipo de gestor de recursos heterogéneo capaz de repartir los recursos entre un entorno transaccional y un entorno Grid dentro del mismo servidor. / Using different applications inside the same environment, heterogeneous environment, is getting more and more usual due the incorporation of the virtualization inside servers. Sharing a server offer advantages in different levels: energy, space, management. Virtualization helps to separate different applications or environments. On the other hand, resource managers have as principal issue offer Quality of Service for different applications, environments or workloads. A streaming server and a CPU intensive application would not collide; the resources they need are different. However, two applications that need CPU processing power will collide.Our proposal offers the possibility to introduce inside the resource manager the capacity to predict these environments. We will work with transactional and Grid environments, and we will increase the QoS and the performance. We need to use simulation techniques for our predictions because a large number of times the system won't be able to be modelled with analytic techniques, for being a saturated system or having features that are hard to reproduce.Simulation is a technique used to predict the behaviour of multiple systems in a large number of areas. Hardware simulations are very common because the building/testing cost of the simulated system (processor, memory, cache,...) is high. However, using simulation in complex environments, as the middleware, and its use in resource management is low. We propose light simulations that can obtain results that can be used in these environments.We will enumerate our contributions: (i) Use simulations to increase the performance and the QoS of those systems. (ii) Improve a global monitoring system for mixed applications (JAVA and C) that gives us information about what happens in the middleware and in the system. (iii) Build a resource manager that can share the resources in a heterogeneous environment an use the prediction to ensure the different QoS parameters that we provide.In the thesis we show how we built the different simulators, the different tools to obtain information and monitorize the applications, and finally the autonomic mechanisms that can feed with the prediction to obtain better results. Results obtained, with great success in the case of the resource manager created for Globus, show and demonstrate that the applied methods in this thesis are suitable to create intelligent resource managers, fed with predictions of the system to take decisions. Finally, we add the built simulations inside a heterogeneous resource manager that shares resources between a transactional environment and a Grid environment inside the same server.

Motor cortex, basal ganglia (BG), and thalamus are anatomically arranged in a recurrent loop whose activity is hypothesized to be involved in the selection of motor actions. Direct (dSPN) and indirect (iSPN) striatal projection neurons receive excitatory input from cortex, and are thought to oppositely modulate cortical activity via BG output to thalamus. Here, we test the central tenets of this model in head-restrained mice performing an operant conditioning task using optogenetic manipulation of dSPNs and iSPNs to determine the effects of activity in each pathway on primary motor cortex. We find that dSPN and iSPN activation has bidirectional, robust, and rapid effects on motor cortex that are highly context-dependent, with distinct effects of each pathway during quiescent and active periods. Thus, the effects of activity in each pathway are at times antagonistic and consistent with classic models, whereas in other behavioral contexts the two pathways will work in the same direction or have no effect at all. In a separate but related project, we describe a direct projection from the globus pallidus externa (GP), a central nucleus of the BG, to frontal regions of the cerebral cortex (FC), which is not typically included in models of BG function. Two cell types make up the GP-FC projection, distinguished by their electrophysiological properties, cortical projection patterns and expression of choline acyteltransferase (ChAT), a genetic marker for the neurotransmitter acetylcholine. These cholinergic GP cells receive basal ganglia input and bidirectionally modulate firing in FC of awake mice. Since GP-FC cells receive dopamine sensitive inhibition from iSPNs and dSPNs, this circuit reveals a pathway by which neuropsychiatric pharmaceuticals can act in the BG and yet modulate frontal cortices. Together, these two projects expand our understanding of the complexities of basal ganglia circuitry and its interactions with cortex.

Previous electrophysiological studies of the basal ganglia in Parkinson's disease (PD) patients have utilized linear analyses in time-or-frequency domains to characterize neuronal discharge patterns. However, these measures do not fully describe the non-linear features of discharge rates and oscillatory activities of basal ganglia neurons.In this original research, we investigate whether non-linear temporal organizations exist in the inter-spike interval series of neurons recorded in the globus pallidus or the subthalamic nucleus in PD patients undergoing surgery for the implantation of deep brain stimulating electrodes.Our data indicate that in approximately 80% of globus pallidus and subthalamic neurons, the raw inter-spike interval sequences have lower entropy values than those observed after shuffling of the original series. This is the first report establishing non-linear temporal organization as a common feature of neuronal discharge in the basal ganglia of PD patients.

Grid computing is a field of research that combines many computers from distant locations to form one large computing resource. In order to be able to make use of the full potential of such a system there is a need to effectively manage resources on the Grid. There are numerous scheduling systems to perform this management for clusters of computers and a few scheduling systems for the Grid. These systems try for optimality (or close to optimality) with the goals of obtaining good throughput and minimizing job completion time. In this research, we examine issues that we believe have not been tackled in schedulers for the Grid. These issues revolve around the problem of coordinating resources belonging to separate administrative domains and scheduling in this context. In order for grid computing's vision of virtual organizations to be realized to its fullest extent, there is a need to implement and test schedulers that find resources and schedule tasks on them in a manner that is transparent to the user. These resources might be on a different administrative domain altogether and obtaining either resource or user account information on those resources might be difficult. Also, each organization might require their own policies and mechanisms to be enforced. Hence having a centralized scheduler is not feasible due to the pragmatics of the Grid. There are two basic aims to this thesis. The first aim is to design and implement a framework that takes administrative concerns into consideration during scheduling. The aim of the framework is to provide a lightweight, extensible, secure and scalable architecture under which multiple scheduling algorithms can be implemented. Second, we evaluate two prototypical of scheduling algorithms in the context of this framework. Scheduling algorithms are diverse and the applications are varied. Thus no single algorithm can obtain a good mapping for every application. We believe that different scheduling algorithms will be necessary to schedule different types of applications. In order to facilitate development of such algorithms, a framework in which it is easy to integrate other scheduling algorithms is necessary. The framework developed in this project is designed for such extensibility. / Master of Science

Die immer dichtere und schnellere Vernetzung von Rechnern und Rechenzentren über Hochgeschwindigkeitsnetzwerke ermöglicht eine neue Art des wissenschaftlich verteilten Rechnens, bei der geographisch weit auseinanderliegende Rechenkapazitäten zu einer Gesamtheit zusammengefasst werden können. Dieser so entstehende virtuelle Superrechner, der selbst aus mehreren Grossrechnern besteht, kann dazu genutzt werden Probleme zu berechnen, für die die einzelnen Grossrechner zu klein sind. Die Probleme, die numerisch mit heutigen Rechenkapazitäten nicht lösbar sind, erstrecken sich durch sämtliche Gebiete der heutigen Wissenschaft, angefangen von Astrophysik, Molekülphysik, Bioinformatik, Meteorologie, bis hin zur Zahlentheorie und Fluiddynamik um nur einige Gebiete zu nennen.<br /> <br /> Je nach Art der Problemstellung und des Lösungsverfahrens gestalten sich solche "Meta-Berechnungen" mehr oder weniger schwierig. Allgemein kann man sagen, dass solche Berechnungen um so schwerer und auch um so uneffizienter werden, je mehr Kommunikation zwischen den einzelnen Prozessen (oder Prozessoren) herrscht. Dies ist dadurch begründet, dass die Bandbreiten bzw. Latenzzeiten zwischen zwei Prozessoren auf demselben Grossrechner oder Cluster um zwei bis vier Grössenordnungen höher bzw. niedriger liegen als zwischen Prozessoren, welche hunderte von Kilometern entfernt liegen.<br /> <br /> Dennoch bricht nunmehr eine Zeit an, in der es möglich ist Berechnungen auf solch virtuellen Supercomputern auch mit kommunikationsintensiven Programmen durchzuführen. Eine grosse Klasse von kommunikations- und berechnungsintensiven Programmen ist diejenige, die die Lösung von Differentialgleichungen mithilfe von finiten Differenzen zum Inhalt hat. Gerade diese Klasse von Programmen und deren Betrieb in einem virtuellen Superrechner wird in dieser vorliegenden Dissertation behandelt. Methoden zur effizienteren Durchführung von solch verteilten Berechnungen werden entwickelt, analysiert und implementiert. Der Schwerpunkt liegt darin vorhandene, klassische Parallelisierungsalgorithmen zu analysieren und so zu erweitern, dass sie vorhandene Informationen (z.B. verfügbar durch das Globus Toolkit) über Maschinen und Netzwerke zur effizienteren Parallelisierung nutzen. Soweit wir wissen werden solche Zusatzinformationen kaum in relevanten Programmen genutzt, da der Grossteil aller Parallelisierungsalgorithmen implizit für die Ausführung auf Grossrechnern oder Clustern entwickelt wurde. / In order to face the rapidly increasing need for computational resources of various scientific and engineering applications one has to think of new ways to make more efficient use of the worlds current computational resources. In this respect, the growing speed of wide area networks made a new kind of distributed computing possible: Metacomputing or (distributed) Grid computing. This is a rather new and uncharted field in computational science. The rapidly increasing speed of networks even outperforms the average increase of processor speed: Processor speeds double on average each 18 month whereas network bandwidths double every 9 months. Due to this development of local and wide area networks Grid computing will certainly play a key role in the future of parallel computing.<br /> <br /> This type of distributed computing, however, distinguishes from the traditional parallel computing in many ways since it has to deal with many problems not occurring in classical parallel computing. Those problems are for example heterogeneity, authentication and slow networks to mention only a few. Some of those problems, e.g. the allocation of distributed resources along with the providing of information about these resources to the application have been already attacked by the Globus software.<br /> <br /> Unfortunately, as far as we know, hardly any application or middle-ware software takes advantage of this information, since most parallelizing algorithms for finite differencing codes are implicitly designed for single supercomputer or cluster execution. We show that although it is possible to apply classical parallelizing algorithms in a Grid environment, in most cases the observed efficiency of the executed code is very poor.<br /> <br /> In this work we are closing this gap. In our thesis, we will<br /> - show that an execution of classical parallel codes in Grid environments is possible but very slow<br /> - analyze this situation of bad performance, nail down bottlenecks in communication, remove unnecessary overhead and other reasons for low performance<br /> - develop new and advanced algorithms for parallelisation that are aware of a Grid environment in order to generelize the traditional parallelization schemes<br /> - implement and test these new methods, replace and compare with the classical ones - introduce dynamic strategies that automatically adapt the running code to the nature of the underlying Grid environment.<br /> <br /> The higher the performance one can achieve for a single application by manual tuning for a Grid environment, the lower the chance that those changes are widely applicable to other programs. In our analysis as well as in our implementation we tried to keep the balance between high performance and generality. None of our changes directly affect code on the application level which makes our algorithms applicable to a whole class of real world applications.<br /> <br /> The implementation of our work is done within the Cactus framework using the Globus toolkit, since we think that these are the most reliable and advanced programming frameworks for supporting computations in Grid environments. On the other hand, however, we tried to be as general as possible, i.e. all methods and algorithms discussed in this thesis are independent of Cactus or Globus.

Le réseau des ganglions de la base (GB) est un ensemble de structures sous corticales, dont la principale fonction est le contrôle du mouvement volontaire. Le Globus Pallidus (GP), équivalent du GPe chez le primate, est un noyau constitué exclusivement de neurones GABAergiques, qui joue un rôle clé dans le fonctionnement des GB de par ses projections inhibitrices diffuses sur l’ensemble des structures de ce macrocircuit. Bien qu’une diversité neuronale au sein du GP ait été suggérée sur les bases de l’origine embryonnaire, de l’expression de protéines spécifiques ou encore de l’activité électrique des neurones, ces différents paramètres n’ont pas été corrélés de manière claire. Notre premier objectif a donc été de corréler les propriétés membranaires de neurones du GP enregistrés en patch-clamp sur des tranches de cerveau de rat avec l’expression spécifique de deux marqueurs neuronaux : une protéine liant le calcium, la parvalbumine (PV) ou un facteur de transcription, Forkhead Box 2 (FoxP2). Nous avons observé des différences électrophysiologiques significatives entre les neurones PV-positifs et FoxP2-positifs. Ce résultat nous a amené à formuler l’hypothèse qu’ayant des propriétés distinctes, les neurones PV-positifs et FoxP2-positifs pouvaient être connectés de manière différente au sein du réseau des ganglions de la base. Nous avons donc réalisé des expériences de traçage neuronal in vivo afin d’identifier les structures cibles de chaque sous-population. Nous montrons que les neurones PV-positifs projettent sur les structures de sortie des ganglions de la base tandis que les neurones FoxP2-positifs projettent uniquement sur le striatum. Enfin, le GP étant majoritairement composé de neurones PV-positifs, nous avons décidé de manipuler spécifiquement l’activité électrique de cette population in vitro et in vivo grâce à l’optogénétique. Nous présentons des résultats montrant que la modulation de l’activité électrique des neurones PV-positifs modifie le comportement moteur chez l’animal vigile. Nos résultats d’immunohistochimie et d’électrophysiologie in vitro démontrent pour la première fois l’existence d’une diversité neuronale au sein du GP. Nos expériences constituent la première étude du rôle des neurones PV-positifs dans le contrôle du mouvement volontaire. / Globus Pallidus (GP in Rodents; GPe in Primates) which belongs to the indirect pathway of basal ganglia is often, if not always, considered as an homogeneous entity which simply relays striatal information through the subthalamic nucleus, downstream to the output of basal ganglia, the substantia nigra pars reticulata. Prototypical GP neurons are often described as fast-spiking GABAergic cells which express parvalbumin (PV) as a neurochemical marker. However, cellular heterogeneity in GP has been suggested by anatomical, neurochemical, fate mapping analysis and electrophysiological activity in vivo but a clear demonstration of the existence of distinct cell types in GP, which requires by definition correlation of electrophysiological activity with neurochemistry and structure, is still missing. The objective of my PhD was i) to determine if the expression of specific neuronal markers in GP neurons is correlated with specific electrophysiological properties, ii) to understand the function of identified cell types in motor control, in order to prove that neuronal diversity exists and matters in GP. We show that electrical activity and repertoire of ionic channels differ in PV-positive and FoxP2-positive neurons. We demonstrate that PV-positive neurons do project on downstream structures whereas FoxP2-positive neurons exclusively target striatum. We report that manipulating PV-positive neurons using optogenetics induce changes in motor behavior. Thus, our results contribute to highlight the function of GP in motor control.

Les ganglions de la base forment un ensemble de structures sous-corticales connues pour leur implication dans les processus sensori-moteurs, cognitifs et motivationnels. L’objectif de ce travail était d’approfondir le rôle des neurones du noyau subthalamique (NST), des neurones de projections et interneurones cholinergiques du striatum et des neurones du globus pallidus externe (GPe) dans la mise en place et l’exécution d’un comportement dans différents contextes motivationnels. Nous nous sommes intéressés à l’influence de l’effort et de la récompense sur l’activité de ces neurones grâce à une approche comportementale associée à des enregistrements extracellulaires unitaires chez le singe éveillé. L’influence de ces facteurs a été appréhendée dans une tâche visuo-motrice dans laquelle différents niveaux d’effort et de récompense étaient imposés à l’animal. Nos résultats comportementaux ont montré une prise en compte de la valeur des stimuli par les animaux. Les résultats électrophysiologiques obtenus montrent une implication de chacune des populations étudiées dans le traitement des informations relatives à l’effort et à la récompense. Ils suggèrent un rôle des neurones du NST, du striatum et du GPe respectivement dans la mise en place, l’exécution et l’évaluation de l’action sur la base de la valeur subjective de la récompense. Nos résultats apportent des informations nouvelles sur les substrats neurophysiologiques qui sous-tendent les processus motivationnels dans la circuiterie des ganglions de la base. / The basal ganglia form a set of subcortical structures known to be involved in sensorimotor, cognitive and motivational processes. The aim of this work was to study the role of the subthalamic nucleus (STN) neurons, the cholinergic interneurons and the projection neurons of the striatum and the neurons of the external part of the globus pallidus (GPe) in the establishment and the execution of a behavior under different motivational contexts. We examined the influence of effort and reward on the activity of these neurons with a behavioral approach combined with extracellular recordings in awake monkeys. The influence of these factors has been investigated in a visuo-motor task in which different levels of effort and reward were imposed on the animal. Our behavioral results showed a consideration of the value of the visual stimuli by the animals. Electrophysiological results showed an implication of each of the neuronal populations studied in the encoding of force and reward related information. These data suggest a role of STN, striatum and GPe in the establishment, the execution and the update of the benefit of the action based on subjective reward value. Our results bring out new features on the neurophysiological substrates underlying motivational processes in basal ganglia circuitry.