Adrien Lebre

Un grand nombre d'applications scientifiques (biologie, climatologie, ...) utilisent et génèrent des quantités de données qui ne cessent de croître en plus de modes d'accès parallèles qui leur sont particuliers. Aux problématiques " out-of-core " ou " parallel I/O " longuement abordées dans un contexte d'accès local, viennent s'ajouter les onsidérations et les contraintes imposées par un environnement distribué comme l'est une grappe. Plusieurs approches ont été proposées à la communauté scientifique dans le but d'améliorer les performances lors d'accès parallèles à des fichiers distants (systèmes de fichiers ou encore librairies d'entrées/sorties spécialisées). Toutefois, ces solutions intègrent des API plus ou moins lourdes mais surtout spécifiques qui nécessitent une connaissance exacte de chaque subtilité interne au modèle. Ce papier présente la solution " aIOLi " : une librairie transparente et efficace pour les accè...

The emergence of Grids enables the sharing of a wide range of resources for solving large-scale computational and data intensive problems in science, engineering and business. While much has been done to build Grid middleware on top of existing operating systems, little has been done to extend the underlying operating systems for enabling and facilitating Grid computing, for example by embedding important functionalities directly into the operating system. XtreemOS project aims at investigating and proposing new services that should be added to current operating systems to build a Grid infrastructure in a simple way. This approach can be seen to have some advantages over conventional Grid middleware. Therefore, the goals of XtreemOS project are to design, implement, evaluate and distribute an open source Grid operating system which supports Grid applications, and capable of running on a wide range of underlying platforms, from clusters to mobiles. Installed on each participating mac...

Virtualization technologies have been a key element in the adoption of Infrastructure-as-a-Service (IaaS) cloud computing platforms as they radically changed the way in which distributed architectures are exploited. However, a closer look suggests that the way of managing virtual and physical resources still remains relatively static.

Although virtualization technologies have recently gained a lot of interest in Grid computing as they allow flexible resource management, the most common way to exploit grids still relies on dedicated services like resource management systems (RMSs) to get resources at a particular time. To improve resource usage, most of these systems provide a best-effort mode where lowest priority jobs can be executed when resources are idle. This particular mode does not provide any guarantee of service and jobs may be killed at any time by the RMS when the nodes they use are subject to higher priority reservations. This behaviour potentially leads to a huge waste of computation time or at least requires users to deal with checkpoints of their best-effort jobs. In this paper, we present Saline, a generic and non-intrusive framework to manage best-effort jobs at grid level through virtual machines (VMs) usage. We discuss the main challenges concerning the design of such a grid system, focusing on...

De nos jours, la gestion des ressources d'une grappe est effectuée en allouant des tranches de temps aux applications, spécifiées par les utilisateurs et de manière statique. Pour un utilisateur, soit les ressources demandées sont sur-estimées et la grappe est sous-utilisée, soit sous-dimensionnées et ses calculs sont dans la plupart des cas perdus. L'apparition de la virtualisation a apporté une certaine flexibilité quant à la gestion des applications et des ressources des grappes. Cependant, pour optimiser l'utilisation de ces ressource, et libérer les utilisateurs d'estimations hasardeuses, il devient nécessaire d'allouer dynamiquement les ressources en fonction des besoins réels des applications : Être capable de démarrer dynamiquement une application lorsqu'une ressource se libère ou la suspendre lorsque la ressource doit être ré-attribuée. En d'autres termes, être capable de développer un système comparable au changement de contexte sur les ordinate...

ABSTRACT Although virtual machine (VM) placement problem has been an active research area over the past decade, the research community is still looking for an open simulation framework that can simulate in an accurate as well as scalable manner VM operations including live migrations. Existing frameworks, however, leverage a naive migration model that considers neither memory update operations nor resource sharing contention, resulting in an underestimate of both the duration of a live migration and the size of migration traffic. In this paper, we propose a simulation framework of virtualized distributed systems with the first class support of live migration operations. We developed a resource share calculation mechanism for VMs and a live migration model implementing the precopy migration algorithm of Qemu/KVM. We extended a widely used simulation toolkit, SimGrid, which allows users to simulate large-scale distributed systems by using user-friendly programming API. Through experiments, we confirmed that our simulation framework correctly reproduced live migration behaviors of the real world under various conditions. Through a first use case, we also confirmed that it is possible to conduct large-scale simulations of complex virtualized workloads upon hundred thousands of VMs upon thousands of physical machines (PMs).

ABSTRACT This paper describes EXECO, a library that provides easy and efficient control of local or remote, standalone or parallel, processes execution, as well as tools designed for scripting distributed computing experiments on any computing platform. After discussing the EXECO internals, we illustrate its interest by presenting two experiments dealing with virtualization technologies on the Grid'5000 testbed.

... Flavien Quesnel and Adrien L`ebre ASCOLA Research Team Ecole des Mines de Nantes-INRIA, LINA (CNRS UMR 6241) Nantes, France Email ... copy, which aims at minimizing the downtime by copying pages on the des-tination node while the process is still running, before ...

ABSTRACT Although the use of virtual environments provided by cloud computing infrastructures is gaining consensus from the scientific community, running applications in these environments is still far from reaching the maturity of more usual computing facilities such as clusters or grids. Indeed, current solutions for managing virtual environments are mostly based on centralized approaches that barter large-scale concerns such as scalability, reliability and reactivity for simplicity. However, considering current trends about cloud infrastructures in terms of size (larger and larger) and in terms of usage (cross-federation), every large-scale concerns must be addressed as soon as possible to efficiently manage next generation of cloud computing platforms. In this work, we propose to investigate an alternative approach leveraging DIStributed and COoperative mechanisms to manage Virtual EnviRonments autonomicallY (DISCOVERY). This initiative aims at overcoming the main limitations of the traditional server-centric solutions while integrating all mandatory mechanisms into a unified distributed framework. The system we propose to implement, relies on a peer-to-peer model where each agent can efficiently deploy, dynamically schedule and periodically checkpoint the virtual environments they manage. The article introduces the global design of the DISCOVERY proposal and gives a preliminary description of its internals.

ABSTRACT One of the principal goals of cloud computing is the outsourcing of the hosting of data and applications, thus enabling a per-usage model of computation. Data and applications may be packaged in virtual machines (VM), which are themselves hosted by nodes, that is, physical machines. Several frameworks have been designed to manage VMs on pools of physical machines; most of them, however, do not efficiently address a major objective of cloud providers: maximizing system utilization while ensuring the QoS. Several approaches promote virtualization capabilities to improve this trade-off. However, the dynamic scheduling of a large number of VMs as part of a large distributed infrastructure is subject to important and hard scalability problems that become even worse when VM image transfers have to be managed. Consequently, most current frameworks schedule VMs statically using a centralized control strategy. In this article, we present distributed VM scheduler, a framework that enables VMs to be scheduled cooperatively and dynamically in large-scale distributed systems. We describe, in particular, how several VM reconfigurations can be dynamically calculated in parallel and applied simultaneously. Reconfigurations are enabled by partitioning the system (i.e., nodes and VMs) on the fly. Partitions are created with a minimum of resources necessary to find a solution to the reconfiguration problem. Moreover, we propose an algorithm to handle deadlocks that may appear because of the partitioning policy. We have evaluated our prototype through simulations and compared our approach with a centralized one. The results show that our scheduler permits VMs to be reconfigured more efficiently: the time needed to manage thousands of VMs on hundreds of machines is typically reduced to a tenth or less