Auswahl der wissenschaftlichen Literatur zum Thema „Grid5000“

This paper deals with the resolution of the Quadratic 3-dimensional Assignment Problem hereafter referred to as Q3AP. Q3AP is an extension of the well-known Quadratic Assignment Problem (QAP) and of the Axial 3-Assignment Problem (A3AP). It finds its application amongst others in Hybrid Automatic Repeat reQuest (HARQ) error-control mechanism used in wireless communication systems. This problem is computationally NP-hard. As far as we know, the largest Q3AP instance size solved to optimality is 13 whereas practical Q3AP instance size can be of 8, 16, 32 or 64. Sequential exact methods such branch-and-bound or sequential metaheuristics are therefore not suited to solve large size instances for the excessive needed computation time. In this paper, we propose parallel hybrid genetic-based metaheuristics for solving the Q3AP. The parallelism in our methods is of two hierarchical levels. The first level is an insular model where a fixed number of genetic algorithms (GA) evolve independently on separate islands and periodically exchange genetic material. The second level is a parallel transformation of individuals in each GA. Implementation has been done using ParadisEO framework, and the experiments have been performed on GRID5000, the French nation-wide computational grid. The experimental results produced by our method were confronted with those reported in the literature. The optimum or the best so far known solutions have been reached in a reasonable computation time.

CGMS (Crop Growth Monitoring System) developed by JRC is an integrated system to monitor crop behaviour and quantitative crop yield forecast that operates on a European scale. To simulate water balance in the root zone the simulation model CGMS-WOFOST (SUPIT & VAN DER GOOT 2003) is used that is based on water storage routing. This study was performed to assess a possible impact of simplifications of the water storage routing based model on simulated water regime in the soil profile. Results of CGMS-WOFOST are compared with results of a more precise Richards’ equation based model HYDRUS-1D (ŠIMŮNEK et al. 2005). 16 scenarios are simulated using HYDRUS-1D. Each scenario represents a single soil profile presented in the selected cell of GRID50 in the Czech Republic. Geometry of the soil profiles, material (texture) definition, root distributions, measured daily rainfall, calculated daily evaporation from the bare soil surface and transpiration of crop canopy were defined similarly to CGMS-WOFOST inputs according to the data stored in the SGDBE40 database. The soil hydraulic properties corresponding to each soil layer were defined using the class transfer rules (WÖSTEN et al. 1999). The bottom boundary conditions were defined either similarly to CGMS-WOFOST bottom boundary condition as a free drainage or as a constant water level 250 cm below the soil surface to demonstrate a ground water impact on the soil profile water balance. The relative soil moisture (RSM) in the root zone during the vegetation period was calculated to be compared with the similar output from CGMS. The RSM values obtained using HYDRUS-1D are higher than those obtained using CGMS-WOFOST mostly due to higher retention ability of HYDRUS-1D. The reasonably higher RSM values were obtained at the end of simulated period using the HYDRUS-1D for the constant water level 250 cm below the soil surface.

L'administration d'une application est une tâche de plus en plus complexe et coûteuse en ressources humaines et matérielles. Nous nous intéressons dans cette thèse à l’administration dans un contexte de grande échelle. Dans ce contexte particulier, nous disposons généralement de plusieurs entités logicielles qui doivent être déployées et gérées sur une infrastructure matérielle de type grille composée de nombreuses machines géographiquement dispersées. L’administration sur ce type d’infrastructure pose de multiples problèmes d’expressivité liés à la description des éléments à administrer, de performance liés à la charge des processus d'administration et la répartition géographique des sites de la grille, d’hétérogénéité matérielle et logicielle, et de dynamicité (panne, coupure de lien réseau, etc.). Nos contributions portent essentiellement sur les problèmes précédemment cités. Un formalisme de description tenant compte du facteur d’échelle est proposé pour décrire l'infrastructure matérielle et logicielle. Nous proposons la répartition de la charge et la diminution du coût de l’administration en utilisant plusieurs systèmes d’administration et en personnalisant la phase d’installation du déploiement. Enfin nous proposons une gestion de l’hétérogénéité matérielle et logicielle. Le travail de cette thèse s’inscrit dans le cadre du projet TUNe. Nous proposons donc une application et une implantation de ces contributions au système TUNe afin de valider notre approche dans le cadre d'une expérimentation en vraie grandeur
Administration of distributed systems is a task increasingly complex and expensive. It consists in to carry out two main activities : deployment and management of application in the process of running. The activity of the deployment is subdivided into several activities : description of hardware and software, conguration, installation and starting the application. This thesis work focuses on large-scale administration which consists to deploy and manage a distributed legacy application composed of several thousands of software entities on physical infrastructure grid made up of hundreds or thousands of machines. The administration of this type of infrastructure creates many problems of expressiveness, performance, heterogeneity and dynamicity (breakdown of machine, network, ...). These problems are generally caused to the scale and geographical distribution of the sites (set of clusters). This thesis contributes to resolve the problems previously cited. Therefore, we propose higher-level descriptions formalisms to describe the structure of hardware and software infrastructures. To reduce the load and increase the performance of the administration, we propose to distribute the deployment system in a hierarchical way in order to distribute the load. The work of this thesis comes the scope of the TUNe (autonomic management system) project. Therefore, we propose to hierarchize TUNe in order to adapt in the context of large-scale administration. We showhow to describe the hierarchy of systems. We also show how to take into account the specicity of the hardware infrastructure at the time of deployment notably the topology, characteristics and types of machine. We dene a process langage allowing to describe the process installation which allow managers to dene thier own installation constraints, according to their needs and preferences. We explore the management of heterogeneity during deployment. Finaly our prototype is validated by an implementation and in the context of a real experimentation

Este estudo apresenta o VCube-Sync, um sistema que utiliza de uma topologia de hipercubos virtuais como base para replicação de um data-store baseado em Tipos de Dados Replicados e Livres de Conflitos - CRDT (Conflictfree Replicated Data Types). Os CRDTs podem garantir a consistência de forma determinística e livre de conflitos. Ao mesmo tempo, hipercubos já foram empregados anteriormente como rede de sobreposição estruturada para a distribuição de mensagens devido à tolerância a falhas e latência logarítmica, permitindo ainda o desenvolvimento de heurísticas de otimização baseadas no conhecimento da configuração da sobreposição. O protocolo de replicação apresentado neste estudo foi baseado no VCube-PS explorando sinergias entre sistemas publicação-subscrição e de replicação. O protocolo foi testado sob várias distribuições de carga e rede usando o testbed Grid5000, e os resultados foram comparados com os de outros protocolos de replicação de pesquisas recentes. Os resultados deste estudo mostram que o VCube-Sync fornece bons resultados em termos de latência, escalabilidade e uso de rede.