Peer-to-peer (P2P) has become a modern distributed computing architecture that supports massively large-scale data management and query processing. Complex query operators such as range operator and join operator are needed by various distributed applications, including content distribution, locality-aware services, computing resource sharing, and many others. This dissertation tackles a number of problems related to range and join query processing in P2P systems: fault-tolerant range query processing under structured P2P architecture, distributed range caching under unstructured P2P architecture, and integration of heterogeneous data under unstructured P2P architecture. To support fault-tolerant range query processing so as to provide strong performance guarantees in the presence of network churn, effective replication schemes are developed at either the overlay network level or the query processing level. To facilitate range query processing, a prefetch-based caching approach is proposed to eliminate the performance bottlenecks incurred by those data items that are not well cached in the network. Finally, a purely decentralized partition-based join query operator is devised to realize bandwidth-efficient join query processing under unstructured P2P architecture. Theoretical analysis and experimental simulations demonstrate the effectiveness of the proposed approaches.

The business model of on-line group buying has been introduced and thrived since 1998 in America. Unfortuately, this internet-based purchase merchanism did not make its growing progress which should be presumed to be an effective and attractive system as an efficient transaction spot on line. Our research specifically focused on the physical consumers' view, including the practical participants in the on-line group buying and the general consumers who never have such experiences on it, to establish a statistical analysis on the effects to consumers' join intention and purchase decision of on-line group buying. We intend to explore the most direct recognition and requirements of the consumers in this commercial mechanism. Also would like to contribute some effective and valuable advices to the future business running under the on-line group buying model for reaching a successful development on it. Over the thorough approach on the collection of related research essays and from the close observation and investigation on the domestic on-line group buying activities, we therefore designed a questionnaire composed with the possible items which might effect the consumers' intention to decide if they would like to join the group buying. By applying the exploratory factor analysis, in this essay, we conclude the 8 major dimensions to our research topic: (1) Familiarity with the group-buying platform and initiators, (2) Recognition to essentials of products, (3) Fulfillment of the cumulative volume of orders, (4) Delivery and payment terms, (5) Convenience to consumers' decision process, (6) Gap between reference prices, (7) Stimulative function, (8) Brand recognition. Furthermore, we can obtain the following inductive reasoning comments from our descriptions and definitions on the 8 major dimensions of our survey. A. Maturity of the internet use can effectively increase consumers' intention to join the on-line group buying. B. Mutation of the modren lifestyle can just meet and satisfy the requirements from consumers. C. Model of group buying can easily and promptly raise the consumers' involvement and understanding on target products.

Since recent years, while issues of environmental protection and green industries have been discussed globally, Taiwan¡¦s agricultural industry has attracted much attention. The local and central governments alike began to promote many agri-facilitating strategies and measures, such as agri-tourism, green technology, and renovation of rural villages. However, on the other hand, the farming population keep on declining through the years, as percentage which youngsters choose farming as an occupation remain significantly low. The current research explored the contemporary Taiwanese college student¡¦s perception of farm living, cognition of agricultural industries, their intentions to choose farming career in various situations, and factors influencing their occupation decisions. The theoretic constructs involved in the present study included: perception of rural living, attitude toward agriculture, knowledge of agricultural industries, values of occupation and intention to join farming. In addition, family supports and understanding of government¡¦s promotion for farming were included also as two practical considerations. The valid empirical sample of the study are 543 college students randomly collected from five public or private universities in southern Taiwan. Data analysis revealed that the reliability and validity of the measuring scales all reached adequate standards, showing good quality of the data. In the research results, regarding the interactive relationship among the main variables, most of the hypothesis obtained supports from the data with statistically significance. Yet, only a few clues could be seen vis-à-vis the higher level of variable relationships, due to the limitation of the applied analysis tool of regression.

<p>Data warehouse maintenance and maintenance cost has been well studied in the literature. Integrating data sources, in a data warehouse environment, may often need data cleaning, transformation, or any other function applied to the data in order to integrate it. The impact on view maintenance, when data is integrated with other comparison operators than defined in theta join, has, however, not been closely looked at in previous studies.</p><p>In this study the impact of using a complex join in data warehouse environment is analyzed to measure how different maintenance strategies are affected when data needs to be integrated using other comparison operators than defined in a theta join. The analysis shows that maintenance cost is greatly increased when using complex joins since such joins often lack optimization techniques which are available when using a theta join. The study shows, among other things, that the join aware capability of sources is not of importance when performing complex joins, and incremental view maintenance is better approach than using recomputed view maintenance, when using complex joins. Strategies for maintaining data warehouses when data is integrated using a complex join are therefore different than when a theta join is used, and different maintenance strategies need to be applied.</p>

Storage services form the core of modern Internet-based services spanning commercial, entertainment, and social-networking sectors. High availability is crucial for these services as even an hour of unavailability can cost them millions of dollars in lost revenue. Unfortunately, it is difficult to build highly available storage services that provide useful correctness properties. Both benign (system crashes, power out- ages etc.) and Byzantine faults (memory or disk corruption, software or configuration errors etc.) plague the availability of these services. Furthermore, the goal of high availability conflicts with our desire to provide good performance and strong correctness guarantees. For example, the Consistency, Availability, and Partition- resilience (CAP) theorem states that a storage service that must be available despite network partitions cannot enforce strong consistency. Similarly, the tradeoff between latency and durability dictates that a low-latency service cannot ensure durability in the presence of data-center wide failures. This dissertation explores the theoretical and practical limits of storage services that can be safe and live despite the presence of benign and Byzantine faults. On the practical front, we use cloud storage as a deployment model to build Depot, a highly available storage service that addresses the above challenges. Depot minimizes the trust clients have to put in the third party storage provider. As a result, Depot clients can continue functioning despite benign or Byzantine faults of the cloud servers. Yet, Depot provides stronger availability, durability, and consistency properties than those provided by many of the existing cloud deployments, without incurring prohibitive performance cost. For example, in contrast to Amazon S3’s eventual consistency, Depot provides a variation of causal consistency on each volume, while tolerating Byzantine faults. On the theoretical front, we explore the consistency-availability tradeoffs. Tradeoffs between consistency and availability have proved useful for designers in deciding how much to strengthen consistency if high availability is desired or how much to compromise availability if strong consistency is essential. We explore the limits of such tradeoffs by attempting to answer the question: What are the semantics that can be implemented without compromising availability? In this work, we investigate this question for both fail-stop and Byzantine failure models. An immediate benefit of answering this question is that we can compare and contrast the consistency provided by Depot with that achievable by an optimal implementation. More crucially, this result complements the CAP theorem. While, the CAP theorem defines a set of properties that cannot be achieved, this work identifies the limits of properties that can be achieved. / text

No description available.

University, Diss., 2000--Mannheim.

Esta Dissertação de Mestrado propõe dois novos algoritmos para tornar mais eficiente o escalonamento on-line de tarefas com dependências estritas em agregados de computadores que usam como middleware para troca de mensagens alguma implementação da MPI (até a versão 2.1). Esses algoritmos foram projetados tendo-se em vista programas construídos no modelo de programação fork/join, onde a operação de fork é usada sobre uma chamada recursiva da função. São eles: 1. O algoritmo RatMD, implementado através de uma biblioteca de primitivas do tipo map-reduce, que funciona para qualquer implementação MPI, com qualquer versão da norma. Utilizado para minimizar o tempo de execução de uma computação paralela; e 2. O algoritmo RtMPD, implementado através de um sistema distribuído sobre daemons gerenciadores de processos criados dinamicamente com a implementação MPICH2 (que implementa a MPI-2). Utilizado para permitir execuções de instâncias maiores de programas paralelos dinâmicos. Ambos se baseiam em roubo de tarefas, que é a estratégia de balanceamento de carga mais difundida na literatura. Para ambos os algoritmos apresenta-se modelagem téorica de custos. Resultados experimentais obtidos ficam dentro dos limites teóricos calculados. RatMD provê uma redução no tempo de execução de até 80% em relação ao algoritmo usual (baseado em round-robin), com manutenção do speedup próximo ao linear e complexidade espacial idêntica à popular implementação com round-robin. RtMPD mantém, no mínimo, o mesmo desempenho que a implementação canônica do escalonamento em MPICH2, dobrando-se o limite físico de processos executados simultaneamente por cada nó. / This Master’s Dissertation proposes two new algorithms for improvement on on-line scheduling of dynamic-created tasks with strict dependencies on clusters of computers using MPI (up to version 2.1) as its middleware for message-passing communication. These algorithms were built targeting programs written on the fork-join model, where the fork operation is always called over an recursive function call. They are: 1. RatMD, implemented as a map-reduce library working for any MPI implementation, on whatever norm’s version. Used for performance gain; and 2. RtMPD, implemented as a distributed system over dynamic-generated processes manager daemons with MPICH2 implentation of MPI. Used for executing larger instances of dynamic parallel programs. Both algorithms are based on the (literature consolidated) work stealing technique and have formal guarantees on its execution time and load balancing. Experimental results are within theoretical bounds. RatMD shows an improvement on the performance up to 80% when paired with more usual algorithms (based on round-robin strategy). It also provides near-linear speedup and just about the same space-complexity on similar implementations. RtMPD keeps, at minimum, the very same performance of the canonical MPICH2 implementation, near doubling the physical limit of simultaneous program execution per cluster node.

Esta Dissertação de Mestrado propõe dois novos algoritmos para tornar mais eficiente o escalonamento on-line de tarefas com dependências estritas em agregados de computadores que usam como middleware para troca de mensagens alguma implementação da MPI (até a versão 2.1). Esses algoritmos foram projetados tendo-se em vista programas construídos no modelo de programação fork/join, onde a operação de fork é usada sobre uma chamada recursiva da função. São eles: 1. O algoritmo RatMD, implementado através de uma biblioteca de primitivas do tipo map-reduce, que funciona para qualquer implementação MPI, com qualquer versão da norma. Utilizado para minimizar o tempo de execução de uma computação paralela; e 2. O algoritmo RtMPD, implementado através de um sistema distribuído sobre daemons gerenciadores de processos criados dinamicamente com a implementação MPICH2 (que implementa a MPI-2). Utilizado para permitir execuções de instâncias maiores de programas paralelos dinâmicos. Ambos se baseiam em roubo de tarefas, que é a estratégia de balanceamento de carga mais difundida na literatura. Para ambos os algoritmos apresenta-se modelagem téorica de custos. Resultados experimentais obtidos ficam dentro dos limites teóricos calculados. RatMD provê uma redução no tempo de execução de até 80% em relação ao algoritmo usual (baseado em round-robin), com manutenção do speedup próximo ao linear e complexidade espacial idêntica à popular implementação com round-robin. RtMPD mantém, no mínimo, o mesmo desempenho que a implementação canônica do escalonamento em MPICH2, dobrando-se o limite físico de processos executados simultaneamente por cada nó. / This Master’s Dissertation proposes two new algorithms for improvement on on-line scheduling of dynamic-created tasks with strict dependencies on clusters of computers using MPI (up to version 2.1) as its middleware for message-passing communication. These algorithms were built targeting programs written on the fork-join model, where the fork operation is always called over an recursive function call. They are: 1. RatMD, implemented as a map-reduce library working for any MPI implementation, on whatever norm’s version. Used for performance gain; and 2. RtMPD, implemented as a distributed system over dynamic-generated processes manager daemons with MPICH2 implentation of MPI. Used for executing larger instances of dynamic parallel programs. Both algorithms are based on the (literature consolidated) work stealing technique and have formal guarantees on its execution time and load balancing. Experimental results are within theoretical bounds. RatMD shows an improvement on the performance up to 80% when paired with more usual algorithms (based on round-robin strategy). It also provides near-linear speedup and just about the same space-complexity on similar implementations. RtMPD keeps, at minimum, the very same performance of the canonical MPICH2 implementation, near doubling the physical limit of simultaneous program execution per cluster node.

Introdução: As doenças articulares, frequentes entre idosos, estão relacionadas ao comprometimento funcional. Objetivo: Avaliar o impacto da doença articular referida no comprometimento funcional em idosos residentes na área urbana do município de São Paulo no período entre 2000 e 2006. Métodos: Este trabalho é parte do Estudo Saúde, Bem-Estar Envelhecimento (SABE). Foram estudados 336 idosos residentes na zona urbana do município de São de Paulo, entrevistados em 2000 e em 2006, representando 162.913 idosos do município. A variável dependente do estudo foi o desenvolvimento de comprometimento funcional e a variável independente foi a doença articular referida entre 2000 e 2006, adicionalmente com as covariáveis sociodemográficas e de saúde. A relação entre as variáveis foi analisada pelo teste de Rao Scot, seguida pelo modelo de regressão logística multinomial e os cálculos do risco atribuível. Considerou-se nível de significância de 5 por cento e intervalos de confiança de 95 por cento . Resultados: No período de 6 anos, observou-se uma incidência de 47,8 por cento de dificuldade de mobilidade e 7,3 por cento de desempenhar atividades de vida diária (AVDs). A doença articular não foi associada ao comprometimento funcional. O sexo feminino foi associado à dificuldade de mobilidade e a idade avançada (70 anos ou mais) e a baixa escolaridade (até 3 anos de estudo) foram associados à dificuldade de desempenho das AVDs. Na população, 9,2 por cento do comprometimento funcional foi atribuído à doença articular. Ainda, entre os idosos que relataram a doença articular, 30,1 por cento do comprometimento funcional foi atribuído a ela. Conclusões: Apesar do impacto da doença articular no comprometimento funcional da população ser representativo, a doença articular não foi associada à incidência de comprometimento funcional nesse estudo / Introduction: The joint disease, common among the elderly, are related to functional impairment. Objective: To evaluate the impact of referred joint disease to functional impairment among elderly living in the urban area of São Paulo between 2000 and 2006. Methods: This study is part of Health Well Being Aging Study (SABE study). We studied 336 elderly residents in the urban area of São Paulo, who were interviewed in 2000 and 2006, representing 162,913 seniors. The dependent variable was functional impairment and the independent variable was joint disease development between 2000 and 2006, in addition to the sociodemographic and health covariates. The association between variables was analyzed using Rao Scot test, followed by multinomial logistic regression model and attributable risk calculations. It was considered a significance level of 5 per cent and confidence interval of 95 per cent . Results: During the 6-year-period, the incidence of mobility impairment was 47.8 per cent and incidence of activities of daily living (ADLs) disability was 7.3 per cent . The joint disease was not associated with functional impairment. Being female was associated with difficulty in mobility difficulties and advanced age (up to 70 years old) and lower education (up to 3 years) were associated with difficulty in performing ADLs. Among population, 9.2 per cent of functional impairment was attributed to joint disease. Still, among the elderly who reported joint disease, 30.1 per cent of functional impairment was attributed to it. Conclusions: despite the expressive impact of joint disease on functional impairment, it was not associated with incidence of disability in this study.