Les machines multicœurs actuelles utilisent une architecture à Accès Mémoire Non-Uniforme (Non-Uniform Memory Access - NUMA). Dans ces machines, les cœurs sont regroupés en nœuds. Chaque nœud possède son propre contrôleur mémoire et est relié aux autres nœuds via des liens d'interconnexion. Utiliser ces architectures à leur pleine capacité est difficile : il faut notamment veiller à éviter les accès distants (i.e., les accès d'un nœud vers un autre nœud) et la congestion sur les bus mémoire et les liens d'interconnexion. L'optimisation de performance sur une machine NUMA peut se faire de deux manières : en implantant des optimisations ad-hoc au sein des applications ou de manière automatique en utilisant des heuristiques. Cependant, les outils existants fournissent trop peu d'informations pour pouvoir implanter efficacement des optimisations et les heuristiques existantes ne permettent pas d'éviter les problèmes de congestion. Cette thèse résout ces deux problèmes. Dans un premier temps nous présentons MemProf, le premier outil d'analyse permettant d'implanter efficacement des optimisations NUMA au sein d'applications. Pour ce faire, MemProf construit des flots d'interactions entre threads et objets. Nous évaluons MemProf sur 3 machines NUMA et montrons que les optimisations trouvées grâce à MemProf permettent d'obtenir des gains de performance significatifs (jusqu'à 2.6x) et sont très simples à implanter (moins de 10 lignes de code). Dans un second temps, nous présentons Carrefour, un algorithme de gestion de la mémoire pour machines NUMA. Contrairement aux heuristiques existantes, Carrefour se concentre sur la réduction de la congestion sur les machines NUMA. Carrefour permet d'obtenir des gains de performance significatifs (jusqu'à 3.3x) et est toujours plus performant que les heuristiques existantes. / Modern multicore systems are based on a Non-Uniform Memory Access (NUMA) design. In a NUMA system, cores are grouped in a set of nodes. Each node has a memory controller and is interconnected with other nodes using high speed interconnect links. Efficiently exploiting such architectures is notoriously complex for programmers. Two key objectives on NUMA multicore machines are to limit as much as possible the number of remote memory accesses (i.e., accesses from a node to another node) and to avoid contention on memory controllers and interconnect links. These objectives can be achieved by implementing application-level optimizations or by implementing application-agnostic heuristics. However, in many cases, existing profilers do not provide enough information to help programmers implement application-level optimizations and existing application-agnostic heuristics fail to address contention issues. The contributions of this thesis are twofold. First we present MemProf, a profiler that allows programmers to choose and implement efficient application-level optimizations for NUMA systems. MemProf builds temporal flows of interactions between threads and objects, which help programmers understand why and which memory objects are accessed remotely. We evaluate MemProf on Linux on three different machines. We show how MemProf helps us choose and implement efficient optimizations, unlike existing profilers. These optimizations provide significant performance gains (up to 2.6x), while requiring very lightweight modifications (10 lines of code or less). Then we present Carrefour, an application-agnostic memory management algorithm. Contrarily to existing heuristics, Carrefour focuses on traffic contention on memory controllers and interconnect links. Carrefour provides significant performance gains (up to 3.3x) and always performs better than existing heuristics.

This document is a master’s thesis analyzing an automated fuel pump management system. In the first chapters we take a general overview of the system and the main problems that we will be facing during the planning and design phases of the project. Later we propose and analyze possible solutions for these problems. Technical system information is revealed in the later chapters. Functional and non functional requirements are discussed along with the main UML diagrams. The research phase of the thesis provides detailed system analysis and software quality reports which later are used to create proposed changes to the system. These changes are analyzed, designed and coded in the final - experimental part of the thesis. The main proposal is to convert the system architecture from flow driven to event driven. This change helps to solve several uncovered architectural problems as well as improve the general quality of the system. These changes are tested and analyzed in the experimental chapter of the thesis and finally the conclusions are made. The main conclusion is that the proposed architectural changes were chosen correctly. This is also supported by the experimental data.

O crescimento explosivo no número de usuários de Internet levou arquitetos de software a reavaliarem questões relacionadas à escalabilidade de serviços que são disponibilizados em larga escala. Projetar arquiteturas de software que não apresentem degradação no desempenho com o aumento no número de acessos concorrentes ainda é um desafio. Neste trabalho, investigamos o impacto do sistema operacional em questões relacionadas ao desempenho, paralelização e escalabilidade de jogos interativos multi-usuários. Em particular, estudamos e estendemos o jogo interativo, multi-usuário, QuakeWorld, disponibilizado publicamente pela id Software sob a licença GPL. Criamos um modelo de paralelismo para a simulação distribuída realizada pelo jogo e o implementamos no servidor do QuakeWorld com adaptações que permitem que o sistema operacional gerencie de forma adequada a execução da carga de trabalho gerada. / The explosive growth in the number of Internet users made software architects reevaluate issues related to the scalability of services deployed on a large scale. It is still challenging to design software architectures that do not experience performance degradation when the concurrent access increases. In this work, we investigate the impact of the operating system in issues related to performance, parallelization, and scalability of interactive multiplayer games. Particularly, we study and extend the interactive, multiplayer game QuakeWorld, made publicly available by id Software under GPL license. We have created a new parallelization model for Quake\'s distributed simulation and implemented that model in QuakeWorld server with adaptations that allows the operating system to manage the execution of the generated workload in a more convenient way.

L’utilisation de concepts avancés de programmation concurrente permet de dépasser les inconvénients de l’utilisation de techniques de bas niveau à base de verrous ou de moniteurs. Elle augmente le niveau d’abstraction, libérant les programmeurs d’applications concurrentes d’une focalisation excessive sur des détails. Cependant, avec les approches actuelles, la logique nécessaire à la mise en place de schémas de coordinations complexes est fragmentée en plusieurs points de l’application sous forme de« join patterns », de notifications et de la logique applicative qui crée implicitement des dépendances entre les canaux de communication et donc, indirectement, les « join patterns » (qui définissent ces canaux). Nous présentons JEScala, un langage qui capture les schémas de coordination (d’une application concurrente) d’une manière plus expressive et modulaire, en s’appuyant sur l’intégration fine d’un système d’évènements avancé et des « join patterns ». Nous implémentons des automates finis à partir de « joins » à l’aide de JEScala et introduisons un langage dédié à la définition de ces automates finis permettant d’en obtenir des implémentations plus efficaces. Nous validons notre approche avec des études de cas et évaluons l’efficacité de son exécution. Nous comparons la performance de trois implémentations d’un automate fini. Nous validons enfin l’idée d’un moniteur d’évènements en créant un programme JEScala concurrent à partir d’un découpage d’un programme séquentiel. / The advanced concurrency abstractions provided by the Join calculus overcome the drawbacks of low-level techniques such as locks and monitors. They rise the level of abstraction, freeing programmers that implement concurrent applications from the burden of concentrating on low-level details. However, with current approaches the coordination logic involved incomplex coordination schemas is fragmented into several pieces including join patterns, data emissions triggered in different places of the application, and the application logic that implicitly creates dependencies among channels, hence indirectly among join patterns. We present JEScala, a language that captures coordination schemas in a more expressive and modular way by leveraging a seamless integration of an advanced event system with join abstractions. We implement Joins-based state machines using JEScala and introduce a domain specific language for finite state machines that make faster alternative implementations possible. We validate our approach with case studies and we provide a first performance assessment. We compare the performance of three different implementations of a finite state machine. Finally, we validate the idea of constructing a concurrent JEScala program by using the parts of a sequential Event-Based program in combination with an event monitor, a component that synchronizes handling of multiple events.

O crescimento explosivo no número de usuários de Internet levou arquitetos de software a reavaliarem questões relacionadas à escalabilidade de serviços que são disponibilizados em larga escala. Projetar arquiteturas de software que não apresentem degradação no desempenho com o aumento no número de acessos concorrentes ainda é um desafio. Neste trabalho, investigamos o impacto do sistema operacional em questões relacionadas ao desempenho, paralelização e escalabilidade de jogos interativos multi-usuários. Em particular, estudamos e estendemos o jogo interativo, multi-usuário, QuakeWorld, disponibilizado publicamente pela id Software sob a licença GPL. Criamos um modelo de paralelismo para a simulação distribuída realizada pelo jogo e o implementamos no servidor do QuakeWorld com adaptações que permitem que o sistema operacional gerencie de forma adequada a execução da carga de trabalho gerada. / The explosive growth in the number of Internet users made software architects reevaluate issues related to the scalability of services deployed on a large scale. It is still challenging to design software architectures that do not experience performance degradation when the concurrent access increases. In this work, we investigate the impact of the operating system in issues related to performance, parallelization, and scalability of interactive multiplayer games. Particularly, we study and extend the interactive, multiplayer game QuakeWorld, made publicly available by id Software under GPL license. We have created a new parallelization model for Quake\'s distributed simulation and implemented that model in QuakeWorld server with adaptations that allows the operating system to manage the execution of the generated workload in a more convenient way.

Cette thèse traite des performances des serveurs de données en multi-cœur. Plus précisément nous nous intéressons au passage à l'échelle avec le nombre de cœurs. Dans un premier temps, nous étudions le fonctionnement interne d'un support d'exécution événementiel multi-cœur. Nous montrons tout d'abord que le faux-partage ainsi que les mécanismes de communications inter-cœurs dégradent fortement les performances et empêchent le passage à l'échelle des applications. Nous proposons alors plusieurs optimisations pour pallier ces comportements. Dans un second temps, nous comparons les performances en multi-cœur de trois serveurs Web chacun représentatif d'un modèle de programmation. Nous remarquons que les différences de performances observées entre les serveurs varient lorsque le nombre de cœurs augmente. Après une analyse approfondie des performances observées, nous identifions la cause de la limitation du passage à l'échelle des serveurs étudiés. Nous présentons une proposition ainsi qu'un ensemble de pistes pour lever cette limitation. / This thesis studies the performances of data servers on multicores. More precisely, we focus on the scalability with the number of cores. First, we study the internals of an event-driven multicore runtime. We demonstrate that false sharing and inter-core communications hurt performances badly, and prevent applications from scaling. We then propose several optimisations to fix these issues. In a second part, we compare the multicore performances of three Webservers, each reprensentative of a programming model. We observe that the differences between each server's performances vary as the number of cores increases. We are able to pinpoint the cause of the scalability limitation observed. We present one approach and some perspectives to overcome this limit.