Thesis (Ph. D.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

La connaissance de la structure tridimensionnelle (3D) des protéines est une information capitale. Néanmoins, le nombre de protéines dont la structure 3D a été déterminée expérimentalement est cent fois plus faible que le nombre de protéines connues aujourd'hui. Cet écart ne pourra pas être comblé, car les techniques expérimentales de détermination de structure (diffraction de rayons X et résonance magnétique nucléaire) sont coûteuses et lentes (un an de travail en moyenne pour une seule protéine). Un moyen d'obtenir plus rapidement la structure 3D de protéines est de la prédire par des moyens bioinformatiques. La technique de prédiction la plus précise actuellement est la modélisation par homologie. Celle-ci est basée sur la similitude de structure entre deux protéines de séquences similaires. L'étape critique de cette méthode est l'étape d'alignement entre la séquence à modéliser et une séquence similaire de structure connue. Notre travail a consisté tout d'abord en la conception d'une nouvelle méthode d'alignement pairé très fiable. Cette méthode a ensuite été incluse dans un système automatique de modélisation par homologie: la bonne qualité des structures prédites par le système trouve en partie son origine dans le programme d'alignement utilisé. Enfin, nous avons appliqué notre système de modélisation automatique à la modélisation de toutes les protéines déduites du génome d'une bactérie pathogène étudiée dans notre unité de recherche: Brucella melitensis. Cela nous a conduit à créer une banque de données structurales et fonctionnelles consacrée au génome de cette bactérie. Cette banque de données est devenue un outil de travail indispensable pour plusieurs équipes de recherche européennes qui étudient Brucella melitensis.

Real-time systems comprise computers that must generate correct results in a timely manner. This involves a wide spectrum of computing systems found in our everyday life ranging from computers in rockets to our mobile phones. The criticality of producing timely results defines the different types of realtime systems. On one hand, we have the so-called hard real-time systems, where failing to meet deadlines may result in a catastrophe. In this thesis we are, however, concerned with firm and soft real-time systems, where missing deadlines is acceptable at the expense of degraded system performance. The usage of firm and soft real-time systems has increased rapidly during the last years, mainly due to the advent of applications in multimedia, telecommunication, and e-commerce. These systems are typically data-intensive, with the data normally spanning from low-level control data, typically acquired from sensors, to high-level management and business data. In contrast to hard real-time systems, the environments in which firm and soft real-time systems operate in are typically open and highly unpredictable. For example, the workload applied on a web server or base station in telecommunication systems varies according to the needs of the users, which is hard to foresee. In this thesis we are concerned with quality of service (QoS) management of data services for firm and soft real-time systems. The approaches and solutions presented aim at providing a general understanding of how the QoS can be guaranteed according to a given specification, even if the workload varies unpredictably. The QoS specification determines the desired QoS during normal system operation, and the worst-case system performance and convergence rate toward the desired setting in the face of transient overloads. Feedback control theory is used to control QoS since little is known about the workload applied on the system. Using feedback control the difference between the measured QoS and the desired QoS is formed and fed into a controller, which computes a change to the operation of the real-time system. Experimental evaluation shows that using feedback control is highly effective in managing QoS such that a given QoS specification is satisfied. This is a key step toward automatic management of intricate systems providing real-time data services.

Providing the ability to increase or decrease allocated resources on demand as the transactional load varies is essential for database management systems (DBMS) deployed on today's computing platforms, such as the cloud. The need to maintain consistency of the database, at very large scales, while providing high performance and reliability makes elasticity particularly challenging. In this thesis, we exploit data partitioning as a way to provide elastic DBMS scalability. We assert that the flexibility provided by a partitioned, shared-nothing parallel DBMS can be used to implement elasticity. Our idea is to start with a small number of servers that manage all the partitions, and to elastically scale out by dynamically adding new servers and redistributing database partitions among these servers as the load varies. Implementing this approach requires (a) efficient mechanisms for addition/removal of servers and migration of partitions, and (b) policies to efficiently determine the optimal placement of partitions on the given servers as well as plans for partition migration. This thesis presents Elasca, a system that implements both these features in an existing shared-nothing DBMS (namely VoltDB) to provide automatic elastic scalability. Elasca consists of a mechanism for enabling elastic scalability, and a workload-aware optimizer for determining optimal partition placement and migration plans. Our optimizer minimizes computing resources required and balances load effectively without compromising system performance, even in the presence of variations in intensity and skew of the load. The results of our experiments show that Elasca is able to achieve performance close to a fully provisioned system while saving 35% resources on average. Furthermore, Elasca's workload-aware optimizer performs up to 79% less data movement than a greedy approach to resource minimization, and also balance load much more effectively.

No description available.

No description available.

A successful information system is the one that meets its design goals. Expressing these goals and subsequently translating them into a working solution is a major challenge for information systems engineering. This thesis adopts the concepts and techniques from goal-oriented (software) requirements engineering research for conceptual database design, with a focus on data quality issues. Based on a real-world case study, a goal-oriented process is proposed for database requirements analysis and modeling. It spans from analysis of high-level stakeholder goals to detailed design of a conceptual databases schema. This process is then extended specifically for dealing with data quality issues: data of low quality may be detected and corrected by performing various quality assurance activities; to support these activities, the schema needs to be revised by accommodating additional data requirements. The extended process therefore focuses on analyzing and modeling quality assurance data requirements. A quality assurance activity supported by a revised schema may involve manual work, and/or rely on some automatic techniques, which often depend on the specification and enforcement of data quality rules. To address the constraint aspect in conceptual database design, data quality rules are classified according to a number of domain and application independent properties. This classification can be used to guide rule designers and to facilitate building of a rule repository. A quantitative framework is then proposed for measuring and comparing DQ rules according to one of these properties: effectiveness; this framework relies on derivation of formulas that represent the effectiveness of DQ rules under different probabilistic assumptions. A semi-automatic approach is also presented to derive these effectiveness formulas.

Modern persistent storage systems must balance two competing imperatives: they must meet strict application-level performance goals and they must reduce the operating costs. The current techniques of either manual tuning by administrators or by over-provisioning resources are either time-consuming or expensive. Therefore, to reduce the costs of management, automated performance-tuning solutions are needed. To address this need, we develop and evaluate algorithms centered around the key thesis that a holistic semantic-aware view of the application and system is needed for automatically tuning and validating the performance of multi-tier storage systems. We obtain this global system view by leveraging structural and semantic information available at each tier and by making this information available to all tiers. Specifically, we develop two key build- ing blocks: (i) context-awareness, where information about the application structure and semantics is exchanged between the tiers, and (ii) dynamic performance models that use the structure of the system to build lightweight resource-to-performance mappings quickly. We implement a prototype storage system, called Akash, based on commodity components. This prototype enables us to study all above scenarios in a realistic rendering of a modern multi-tier storage system. We also develop a runtime tool, Dena, to analyze the performance and behaviour of multi-tier server systems. We apply these tools and techniques in three real-world scenarios. First, we leverage application context-awareness at the storage server in order to improve the performance of I/O prefetching. Tracking application access patterns per context enables us to improve the prediction accuracy for future access patterns, over existing algorithms, where the high interleaving of I/O accesses from different contexts make access patterns hard to recognize. Second, we build and leverage dynamic performance models for resource allocation, providing consistent and predictable performance, corresponding to pre-determined application goals. We show that our dynamic resource allocation algorithms minimize the interference effects between e-commerce applications sharing a common infrastructure. Third, we introduce a high-level paradigm for interactively validating system performance by the system administrator. The administrator leverages existing performance models and other semantic knowledge about the system in order to discover bottlenecks and other opportunities for performance improvements. Our evaluation shows that our techniques enable significant improvements in performance over current approaches.

This study determined the sensitivity and specificity of questionnaires and administrative databases compared to a clinical reference standard for asthma. 208 schoolchildren from a population-based sample participated. They underwent a physician assessment, spirometry, methacholine challenge, exhaled nitric oxide and skin testing. Data was linked to the Ontario Asthma Surveillance Information System. “Questionnaire diagnosis” was an affirmative response to physician-diagnosed asthma. “Database diagnosis” was 2 outpatient visits or 1 hospitalization within 2 years. “Clinical diagnosis” required a physician assessment and objective findings of asthma. “Questionnaire diagnosis” of asthma was specific (92.1%) but not sensitive (75.3%) compared to the “clinical diagnosis”. “Database diagnosis” was sensitive (87.5%) but not specific (64.8%). Both sources had an excellent negative predictive value (97-98%) but poor positive predictive value (24-55%). Epidemiologic methods accurately identify those without asthma but are poor at identifying those with asthma, leading to an overestimation of asthma prevalence and dilution of risk estimates.

This study determined the sensitivity and specificity of questionnaires and administrative databases compared to a clinical reference standard for asthma. 208 schoolchildren from a population-based sample participated. They underwent a physician assessment, spirometry, methacholine challenge, exhaled nitric oxide and skin testing. Data was linked to the Ontario Asthma Surveillance Information System. “Questionnaire diagnosis” was an affirmative response to physician-diagnosed asthma. “Database diagnosis” was 2 outpatient visits or 1 hospitalization within 2 years. “Clinical diagnosis” required a physician assessment and objective findings of asthma. “Questionnaire diagnosis” of asthma was specific (92.1%) but not sensitive (75.3%) compared to the “clinical diagnosis”. “Database diagnosis” was sensitive (87.5%) but not specific (64.8%). Both sources had an excellent negative predictive value (97-98%) but poor positive predictive value (24-55%). Epidemiologic methods accurately identify those without asthma but are poor at identifying those with asthma, leading to an overestimation of asthma prevalence and dilution of risk estimates.