Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

L'explosion actuelle de l'utilisation des applications en temps réel, telles que la téléphonie et la vidéo haute définition, fait en sorte qu'une demande croissante en bande passante existe. Or, la fibre optique offre un très grand potentiel en termes de bande passante allant physiquement jusqu'à 50 terabits par seconde par fibre optique en utilisant un multiplexage en longueurs d'onde. La commutation de rafales en optique est une nouvelle technique de commutation permettant de tirer profit des avantages de la commutation de paquets et de la commutation de circuits. Une problématique majeure est le taux de perte élevé à cause du problème de la contention. Plusieurs techniques de résolution de la contention existent, notamment la déflexion et la retransmission. Dans la première partie de ce travail, on s'intéresse à combiner dynamiquement la déflexion et la retransmission d'une manière adaptative en tenant compte de l'état du réseau. L'algorithme proposé permet d'effectuer le plus de déflexions possibles tant que celles-ci ne déstabilisent pas le réseau. Les résultats démontrent également, avec des topologies fortement connectées telles que COST239, que l'algorithme proposé permet de diminuer radicalement les pertes en utilisant un ratio de déflexion très grand. La deuxième partie de ce travail porte sur la prévention de la contention en utilisant des tables de routage optimisées. Ces tables de routage sont optimisées en utilisant un modèle graphique probabiliste utilisé en intelligence artificielle, soit un réseau bayésien. ______________________________________________________________________________ MOTS-CLÉS DE L’AUTEUR : Fibre optique, WDM, Commutation de rafales, Contention, Déflexion, Retransmission.

This thesis will explore contentious politics in China and the role of digitally mediated social networks as mobilizing structures for contentious collective action. Drawing on a case study of a group of sixteen high school students in Beijing, this research analyzes their use of Weixin, a multi-functional instant messaging platform, to develop and maintain their social networks, and how the changes to their networks of strong and weak ties may be conducive to mobilization of contention. This study also explores the potential for this communication platform to become a robust counterpublic sphere in which its 355 million users can feel free to express themselves. The findings of this research demonstrate that most users believe Weixin to be a private communication space, populated by trusted ties, with whom they feel free to express themselves. However, suspicions of state surveillance and incidents of censorship have had an impact for wary users. While networks mediated by Weixin are primarily virtual extension of real intermittent networks, users have found this platform to have an impact on increasing their strong ties, and building trust in their relationships. In times of crisis, and in the final decision-shaping process of mobilization, it is these strong ties that make networks so valuable. However, the respondents do not show a great propensity to use Weixin to build a more heterogeneous network that affords them access to a broader range of social groups and information, a necessary precondition for the socialization function of networks in mobilizing contentious collective action. / Graduate / 0615 / cindytse40@hotmail.com

Th.--Chir. dent.--Reims, 1981. N°: 20.

With the increasing use of multimedia applications on the wireless network, the functionalities of the IEEE 802.11 WLAN was extended to allow traffic differentiation so that priority traffic gets quicker service time depending on their Quality of Service (QoS) requirements. The extended functionalities contained in the IEEE Medium Access Control (MAC) and Physical Layer (PHY) Specifications, i.e. the IEEE 802.11e specifications, are recommended values for channel access parameters along traffic lines and the channel access parameters are: the Minimum Contention Window CWmin, Maximum Contention Window CWmax, Arbitration inter-frame space number, (AIFSN) and the Transmission Opportunity (TXOP). These default Enhanced Distributed Channel Access (EDCA) contention values used by each traffic type in accessing the wireless medium are only recommended values which could be adjusted or changed based on the condition of number of associated nodes on the network. In particular, we focus on the Contention Window (CW) parameter and it has been shown that when the number of nodes on the network is small, a smaller value of CWmin should be used for channel access in order to avoid underutilization of channel time and when the number of associated nodes is large, a larger value of CWmin should be used in order to avoid large collisions and retransmissions on the network. Fortunately, allowance was made for these default values to be adjusted or changed but the challenge has been in designing an algorithm that constantly and automatically tunes the CWmin value so that the Access Point (AP) gives out the right CWmin value to be used on the WLAN and this value should be derived based on the level of activity experienced on the network or predefined QoS constraints while considering the dynamic nature of the WLAN. In this thesis, we propose the use of feedback based control and we design a controller for wireless medium access. The controller will give an output which will be the EDCA CWmin value to be used by contending stations/nodes in accessing the medium and this value will be based on current WLAN conditions. We propose the use of feedback control due to its established mathematical concepts particularly for single-input-single-output systems and multi-variable systems which are scenarios that apply to the WLAN.

With the advancement of technology, multi-cores with shared cache have been used in real-time applications. In such systems, some cores run real-time applications and some cores run other non-critical applications that do not have strict deadline. Due to the sharing of cache by multi-core processors, problems predicting the actual execution time and the execution time of real-time applications have emerged. To address these problems, cache memory with prioritized replacement policy is proposed. Most of the work is carried out in high-level hardware designs and software based application level designs. No low-level hardware implementations of cache memory with prioritized replacement circuits have been designed to the best of my knowledge. My thesis focuses on designing a LRU replacement circuit that is prioritized based on the application the processor is running. Real-time applications acquire priority in using the cache memory over other applications which enhance the seamless execution of the real-time application and hence supports execution time predictability which in turn helps improve the potential of multi-core computing of real-time systems. The speed, size and power overhead are analyzed by placing the N-way set associative LRU as a part of cache of size 128KB designed using 65nm CMOS technology.

In this thesis we design and implement a benchmarking tool for applications' sensitivity to main memory bandwidth contention, in a multi-core environment, on an ARM Cortex-A15 CPU. The tool is supposed to minimize usage of shared resources, except for the main memory bandwidth, allowing it to isolate the effects of the bandwidth contention only. The difficulty in doing this lies in using a correct memory access pattern for this purpose, i.e. which memory addresses to access, in which order and at what rate in order to minimize cache usage while generating a high and controllable main memory bandwidth usage. We manage to implement a tool with low cache memory usage while still being able to saturate the main memory bandwidth. The tool uses a proportional-integral controller to control the amount of bandwidth it uses. We then use the tool to investigate the memory behaviour of the platform and of some applications when the tool is using a variable amount of bandwidth. However, we have some difficulties in analyzing the results due to the lack of support for hardware performance counters in the operating system we are using and are forced to rely on hardware timers for our data gathering. Another difficulty is the platform's limited L2 cache bandwidth, which leads to a heavy impact on L2 cache read latency by the tool. Despite this, we are able to draw some conclusions on the bandwidth usage of other applications in optimal cases with the help of the tool.

Lock-based concurrency control suffers from programmability, scalability, and composability challenges. These challenges are exacerbated in emerging multicore architectures, on which improved software performance must be achieved by exposing greater concurrency. Transactional memory (TM) is an emerging alternative synchronization model for shared memory objects that promises to alleviate these difficulties. In this dissertation, we consider software transactional memory (STM) for concurrency control in multicore real-time software, and present a suite of real-time STM contention managers for resolving transactional conflicts. The contention managers are called ECM, RCM, LCM, PNF, and FBLT. RCM and ECM resolve conflicts using fixed and dynamic priorities of real-time tasks, respectively, and are naturally intended to be used with the fixed priority (e.g., G-RMA) and dynamic priority (e.g., G-EDF) multicore real-time schedulers, respectively. LCM resolves conflicts based on task priorities as well as atomic section lengths, and can be used with G-EDF or G-RMA schedulers. Transactions under ECM, RCM, and LCM may retry due to conflicts with higher priority tasks even when there are no shared objects, i.e., transitive retry. PNF avoids transitive retry and optimizes processor usage by lowering the priority of retrying transactions, thereby enabling other non-conflicting transactions to proceed. PNF, however, requires a priori knowledge of all requested objects for each atomic section, which is inconsistent with the semantics of dynamic STM. Moreover, its centralized design increases overhead. FBLT avoids transitive retry, do not require a priori knowledge of requested objects, and has a decentralized design. We establish upper bounds on transactional retry costs and task response times under the contention managers through schedulability analysis. Since ECM and RCM preserve the semantics of the underlying real-time scheduler, their maximum transactional retry cost is double the maximum atomic section length. This is improved in the design of LCM, which achieves shorter retry costs and tighter upper bounds. As PNF avoids transitive retry and improves processor usage, it yields shorter retry costs and tighter upper bounds than ECM, RCM, and LCM. FBLT\'s upper bounds are similarly tight because it combines the advantages of PNF and LCM. We formally compare the proposed contention managers with each other, with lock-free synchronization, and with multiprocessor real-time locking protocols. Our analysis reveals that, for most cases, ECM, RCM, and LCM achieve higher schedulability than lock-free synchronization only when the atomic section length does not exceed half of lock-free synchronization\'s retry loop length. With equal periods and greater access times for shared objects, atomic section length under ECM, RCM, and LCM can be much larger than the retry loop length while still achieving better schedulability. With proper values for LCM\'s design parameters, atomic section length can be larger than the retry loop length for better schedulability. Under PNF, atomic section length can exceed lock-free\'s retry loop length and still achieve better schedulability in certain cases. FBLT achieves equal or better schedulability than lock-free with appropriate values for design parameters. The schedulability advantage of the contention managers over multiprocessor real-time locking protocols such as Global OMLP and RNLP depends upon the value of $s_{max}/L_{max}$, the ratio of the maximum transaction length to the maximum critical section length. FBLT\'s schedulability is equal or better than Global OMLP and RNLP if $s_/L_ le 2$. Checkpointing enables partial roll-back of transactions by recording transaction execution states (i.e., checkpoints) during execution, allowing roll-back to a previous checkpoint instead of transaction start, improving task response time. We extend FBLT with checkpointing and develop CP-FBLT, and identify the conditions under which CP-FBLT achieves equal or better schedulability than FBLT. We implement the contention managers in the Rochester STM framework and conduct experimental studies using a multicore real-time Linux kernel. Our studies reveal that among the contention managers, CP-FBLT has the best average-case performance. CP-FBLT\'s higher performance is due to the fact that PNF\'s and LCM\'s advantages are combined into the design of FBLT, which is the base of CP-FBLT. Moreover, checkpointing improves task response time. The contention managers were also found to have equal or better average-case performance than lock-free synchronization: more jobs meet their deadlines using CP-FBLT, FBLT, and PNF than lock-free synchronization by 34.6%, 28.5%, and 32.4% (on average), respectively. The superiority of the contention managers is directly due to their better conflict resolution policies. Locking protocols such as OMLP and RNLP were found to perform better: more jobs meet their deadlines under OMLP and RNLP than any contention manager by 12.4% and 13.7% (on average), respectively. However, the proposed contention managers have numerous qualitative advantages over locking protocols. Locks do not compose, whereas STM transactions do. To allow multiple objects to be accessed in a critical section, OMLP assigns objects to non-conflicting groups, where each group is protected by a distinct lock. RNLP assumes that objects are accessed in a specific order to prevent deadlocks. In contrast, STM allows multiple objects to be accessed in a transaction in any order, while guaranteeing deadlock-freedom, which significantly increases programmability. Moreover, STM offers platform independence: the proposed contention managers can be entirely implemented in the user-space as a library. In contrast, real-time locking protocols such as OMLP and RNLP must be supported by the underlying platform (i.e., operating system or virtual machine). / Ph. D.

Dans ces dernières années, les Systèmes de Transport Intelligents (STI) ont été considérés comme l'un des domaines de recherche les plus émergents en raison de leur rôle prometteur dans l'amélioration de la gestion du trafic et de la sécurité routière. Les applications coopératives de sécurité, étant les plus cruciales, ont gagné beaucoup d'intérêt. L'efficacité de ces applications dépend largement de l'échange efficace de deux principaux types d'informations. L'information de localisation périodique correspondant à l'information de localisation du voisinage et l'information événementielle qui est transmise en multi-sauts et générée lors de la détection d'une situation d'urgence. En raison de la caractéristique à grande échelle des STI, cette information fait l'objet du problème de congestion dans le réseau. L'objectif de cette thèse est d'assurer un contrôle fiable et robuste des informations de sécurité permettant de réduire la congestion du canal tout en tenant en compte des exigences des applications de sécurité. Nous examinons la diffusion de l'information événementielle en proposant une approche a multi-sauts qui a montré une amélioration de la réception de l'information. Cependant, cette approche reste très sensible à la charge de canal résultant de transmissions de l'information de localisation périodiques. D'autre part, la transmission efficace de l'information événementielle repose essentiellement sur la détection précise des événements de sécurité et en conséquence sur la précision de l'information de localisation. Ainsi, nous proposons un mécanisme de contrôle de l'information de localisation afin de fournir une meilleure précision et limiter la charge du canal. Les approches proposées dans cette thèse ont profondément étudié le compromis entre le respect des exigences des applications de sécurité et la gestion efficace de la congestion dans le réseau véhiculaire. / In the last decades, Intelligent Transportation Systems (ITS) have been considered as one of the most emerging research area due to their promising role in promoting traffic efficiency and enhancing road safety. ITS cooperative safety applications, being the most vital and critical, have gained a lot of attention. The effectiveness of these applications depends widely on the efficient exchange of two main types of information. The periodic awareness corresponding to the one-hop location information of surrounding environment and the multi-hop event-driven information generated at the detection of a safety situation. Due to the large scale characteristic of ITS, this information is expected to be subject to severe congestion which might impact its reliable reception. The goal of this thesis is to focus on the reliable and robust control of safety-related information by reducing the channel congestion and at the same time taking into account the requirements of safety applications. We address first the event-driven safety information. We proposed a multi-hop policy showed to improve the dissemination of the event-driven information. However, it remains strongly sensitive to the channel load resulting from periodic awareness transmissions. On the other hand, the effective transmission of event-driven information depends primarily on the accurate detection of safety events and accordingly on the accuracy of awareness. Thus, we provide an efficient awareness control mechanism in order to provide better accuracy and limit the channel congestion. The approaches proposed in this thesis have deeply investigated the trade-off between ensuring the requirements of cooperative safety and the efficient management of congestion in vehicular network.

No description available.