Este estudo tenciona discutir o problema da conceituação das doenças mentais, a partir dos DSMs e dos diferentes paradigmas que os embasaram. O DSM (manual estatístico e diagnóstico de transtornos mentais) é um manual, de influência internacional, para profissionais da saúde mental, que lista diferentes categorias de transtornos mentais, de acordo com a Associação Psiquiátrica Norte-Americana. Desde a sua primeira publicação, em 1952, já foi submetido a cinco revisões (DSM II, DSM III, DSM III-R, DSM IV e DSM IVTR). Escolhemos a categoria diagnóstica da depressão, objetivando realizar um rastreamento conceitual, desde o DSM II - modelo até então marcado pela psicanálise, depois ressaltando o DSM III, que, em 1980, promove uma mudança de paradigma no conhecimento psiquiátrico, ao apresentar um modelo que se propõe descritivo e ateórico até o DSM IV-TR. Dessa perspectiva, são assinaladas algumas considerações e pontos de discussão entre a chamada psiquiatria biológica e a psicanálise, no que diz respeito às suas respectivas influências na forma de entender o diagnóstico psiquiátrico, enfatizando a categoria diagnóstica da depressão. / This study intends to discuss the issue of categorizing mental disorders, considering DSMs and its paradigms. The Diagnostic and Statistical Manual of Mental Disorder (DSM) is a handbook, used worldwide, for mental professionals that lists different categories of mental disorder, according to the American Psychiatric Association. The DSM has gone through five revisions (DSM II, DSM III, DSM III-R, DSM IV e DSM IV- TR) since it was first published in 1952. Electing the concept of depression, this study analyses its diagnosis criteria from DSM II, a psychoanalytic influenced manual to DSM IV-TR, with an emphasis upon DSM III which, in 1980, represents a turning point in psychiatric paradigms, intending to be not theoretical and descriptive. From this perspective, the intention is to outline some considerations and discussions between the so-called biological psychiatric and the psychoanalytic theory, according to its influences in the psychiatric diagnosis understanding, emphasizing the category of depression.

Este estudo tenciona discutir o problema da conceituação das doenças mentais, a partir dos DSMs e dos diferentes paradigmas que os embasaram. O DSM (manual estatístico e diagnóstico de transtornos mentais) é um manual, de influência internacional, para profissionais da saúde mental, que lista diferentes categorias de transtornos mentais, de acordo com a Associação Psiquiátrica Norte-Americana. Desde a sua primeira publicação, em 1952, já foi submetido a cinco revisões (DSM II, DSM III, DSM III-R, DSM IV e DSM IVTR). Escolhemos a categoria diagnóstica da depressão, objetivando realizar um rastreamento conceitual, desde o DSM II - modelo até então marcado pela psicanálise, depois ressaltando o DSM III, que, em 1980, promove uma mudança de paradigma no conhecimento psiquiátrico, ao apresentar um modelo que se propõe descritivo e ateórico até o DSM IV-TR. Dessa perspectiva, são assinaladas algumas considerações e pontos de discussão entre a chamada psiquiatria biológica e a psicanálise, no que diz respeito às suas respectivas influências na forma de entender o diagnóstico psiquiátrico, enfatizando a categoria diagnóstica da depressão. / This study intends to discuss the issue of categorizing mental disorders, considering DSMs and its paradigms. The Diagnostic and Statistical Manual of Mental Disorder (DSM) is a handbook, used worldwide, for mental professionals that lists different categories of mental disorder, according to the American Psychiatric Association. The DSM has gone through five revisions (DSM II, DSM III, DSM III-R, DSM IV e DSM IV- TR) since it was first published in 1952. Electing the concept of depression, this study analyses its diagnosis criteria from DSM II, a psychoanalytic influenced manual to DSM IV-TR, with an emphasis upon DSM III which, in 1980, represents a turning point in psychiatric paradigms, intending to be not theoretical and descriptive. From this perspective, the intention is to outline some considerations and discussions between the so-called biological psychiatric and the psychoanalytic theory, according to its influences in the psychiatric diagnosis understanding, emphasizing the category of depression.

Existe una creciente demanda de sistemas online, especial-mente de aquellos que requieren procesamiento de informa-ción. Esta demanda sumada a las nuevas tecnologías de monitoreo (como por ejemplo las redes de sensores) impulsaron un nuevo tipo de aplicación que requiere bajas latencias y procesamiento continuo de grandes volúmenes de datos (los cuales arriban en forma de streams). El proce-samiento de streams constituye un paradigma de cómputo relacionado con SIMD que permite que algunos tipos de apli-caciones puedan explotar una forma de procesamiento para-lelo y puede emplearse en diferentes dominios, como por e-jemplo para la implementación de sistemas financieros, moni-toreo basado en sensores, sistemas militares, monitoreo de red, etc. Si bien los sistemas de gestión de bases de datos (DBMS) pueden utilizarse para implementar este tipo de apli-caciones, las restricciones de bajas latencias de procesa-miento y grandes volúmenes de datos a procesar los vuelve inadecuados. Una mejor alternativa son los sistemas de ges-tión de streams de datos, usualmente sistemas distribuidos de gestión de streams de datos (DSMS por su sigla en inglés) de-bido a que estas aplicaciones son inherentemente distribuidas y por lo tanto las soluciones distribuidas son naturales y proveen mejoras en cuanto a escalabilidad y performance. Esta tesis se enfoca en dos aspectos desafiantes pertenecien-tes al campo de los sistemas distribuidos en general y al de los DSMS en particular: (1) tolerancia a fallas capaz de re-sistir fallas a nivel de nodos y de red y (2) gestión de carga en sistemas federados. Nuestro enfoque al problema de la to-lerancia a fallas se basa en replicación capaz de enmascarar tanto las fallas a nivel de los nodos como a nivel de las redes. Nuestroprotocolo, denominado Disponibilidad y Consistencia Ajustable a las Aplicaciones (DCAA) puede manejar adecua-damente la relación entre disponibilidad y consistencia, man-teniendo (si es posible) la disponibilidad especificada por el usuario o la aplicación, pero produciendo (eventualmente) los resultados correctos. Al mismo tiempo, DCAA también trata de producir el menor número de resultados incorrectos (impre-cisos) que luego deberían requerir corrección. La principal diferencia entre DCAA y enfoques previos sobre tolerancia a fallas en el campo de los DSMS es que DCAA soporta al mismo tiempo diferentes restricciones en las aplicaciones, esto quiere decir que cada aplicación puede potencialmente tener distintas preferencias de disponibilidad y consistencia. Por otro lado presentaremos un nuevo protocolo de gestion de carga denominado Mecanismo de Precio Acotado (MPA), el cual permite que nodos autonomos (participantes colabora-tivos) compartan su carga sin la necesidad de contar con recursos suficientes para la operación durante picos de carga. MPA es un protocolo basado en contratos donde cada nodo practica una negociación offline y los participantes migran carga en tiempo de ejecución únicamente a nodos (pares) con los cuales mantienen un contrato (y pagan mutuamente de acuerdo al precio contratado). Este protocolo de gestión de carga ofrece incentivos que promueven la participación de los nodos y produce una buena distribución de carga (a nivel global del sistema). Los aportes mas importantes de nuestro enfoque por sobre trabajos previos basados en economías de cómputo son su estabilidad, predecibilidad, baja carga de procesamiento, privacidad y promoción de relaciones entre participantes, posibilitando que los mismos pueden crear y explotar estas relaciones privilegiadas. El protocolo MPA es general y por lo tanto puede utilizarse para la gestión de carga de cualquier entorno federado y no sólo bajo DSMS. Más aún, este nuevo protocolo de gestión de carga debe no sólo traba-jar en los típicos entornos colaborativos sino que también debe ser capaz de solucionar escenarios más reales, donde cada nodo (probablemente parte de diferentes organizaciones autónomas) juega bajo distintas reglas, tratando de maximi-zar su propia ganancia sin cooperar necesariamente con sus pares. Además de los modelos económicos existen varios tra-bjos basados en SLA (Service Level Agreements) para solucio-nar el problema de la gestión de carga cuando el entorno no es colaborativo. Mostraremos que los modelos SLA no proveen una solucion completa y que los acuerdos entre pares usual-mente proveen mejores resultados. Si bien esta tesis parece tener dos focos en lugar de uno, es importante notar que ata-caremos especialmente el problema de la gestión de carga en sistemas distribuidos federados. La relación entre este enfo-que y la tolerancia a fallas radica en los contratos negocia-dos: además de precio y tareas (carga), los contratos pueden incluir disponibilidad, característica que vuelve especialmente importante la tolerancia a fallas. / There is an increased demand for online systems,especially those requiring information processing. This demand added to new monitoring technologies (like sensors networks) have motivated a new type of application that requires low latency and continuous processing of large volumes of data (arriving as streams). Stream processing is a computer programming paradigm, related to SIMD, that allows some applications to more easily exploit a form of parallel pro-cessing and it can be employed in many different domains, such as financial systems, sensor based monitoring, milita-ry systems, network monitoring, etc. Even when traditional database management systems (DBMS) can be used to handle these applications, the low latency and high volume pro-cessing constrains make them not suitable. A much better alternative are the data stream management systems, usually distributed data stream management systems (DSMS) because these are inherently distributed applications so distributed solutions are natural and providers of scalabi-lity and performance improvements. This thesis focuses on two challenges faced by distributed systems in general and DSMS in particular: (1) fault tolerance able to resist node and network failures and (2) load management in fede-rated systems. The fault tolerance approach is based on re-plication and our protocol can resist most node and net-work failures. It is called Disponibilidad y Consistencia Ajustable a las Aplicaciones (DCAA) and addresses the availability/consistency relation by maintaining (if possi-ble), the availability specified by the user or the appli-cation but (eventually) producing correct results. At the same time, DCAA also attempts to produce the minimum number of incorrect (inaccurate) results that will need correction. The main difference of DCAA over previous approaches for fault tolerance in DSMS is that DCAA supports at the same time different application constrains, this means that each application can potentially choose a different preference of availability and consistency. Our load management protocol, called Mecanismo de Precio Acota-do (MPA) enable autonomous nodes (collaborative participa-nts) to share their load without having the required re-sources for peak load work. MPA is a contract based proto-col where nodes practice an offline negotiation and parti-cipants migrate load at execution time only to peers with whom they hold a contract (and pay each other according to the contracted price). This load management protocol offers incentives that promote participation and produces a good (system wide level) load distribution. The key differences of our approach over previous works based on computational economies are its stability, predictability, lightweight, privacy, and promotion of the relationships among participants, enabling them to create and exploit these privileged relationships. The MPA protocol is gene-ral, so it can be used to manage load in any federated en-vironment, and not only DSMS. Moreover, this new load ma-nagement protocol should not only work under the typical collaborative environment, but also should be able to address the more realistic scenery where each node (proba-bly part of different and autonomous organizations) plays under different rules trying to maximize their own gain, without necessarily cooperating with their partners. Besi-des economic models there are various works based on SLA (service level agreements) to solve the load management problem when the environment is not a collaborative one. We will show that SLA models do not provide a complete solution and that peer agreements usually provide better results. Although this thesis seems to have two focuses instead of one, it is important to notice that we espe-cially address the load management problem under federated distributed systems. The relation among this focus and fault tolerance is in the negotiated contracts: besides price and tasks (load), contracts can include availability, which raises the importance of fault tolerance.

Energy consumption has become a first-class optimization goal in design and implementation of data-intensive computing systems. This is particularly true in the design of database management systems (DBMS), which is one of the most important servers in software stack of modern data centers. Data storage system is one of the essential components of database and has been under many research efforts aiming at reducing its energy consumption. In previous work, dynamic power management (DPM) techniques that make real-time decisions to transition the disks to low-power modes are normally used to save energy in storage systems. In this research, we tackle the limitations of DPM proposals in previous contributions and design a dynamic energy-aware disk storage system in database servers. We introduce a DPM optimization model integrated with model predictive control (MPC) strategy to minimize power consumption of the disk-based storage system while satisfying given performance requirements. It dynamically determines the state of disks and plans for inter-disk data fragment migration to achieve desirable balance between power consumption and query response time. Furthermore, via analyzing our optimization model to identify structural properties of optimal solutions, a fast-solution heuristic DPM algorithm is proposed that can be integrated in large-scale disk storage systems, where finding the most optimal solution might be long, to achieve near-optimal power saving solution within short periods of computational time. The proposed ideas are evaluated through running simulations using extensive set of synthetic workloads. The results show that our solution achieves up to 1.65 times more energy saving while providing up to 1.67 times shorter response time compared to the best existing algorithm in literature. Stream join is a dynamic and expensive database operation that performs join operation in real-time fashion on continuous data streams. Stream joins, also known as window joins, impose high computational time and potentially higher energy consumption compared to other database operations, and thus we also tackle energy-efficiency of stream join processing in this research. Given that there is a strong linear correlation between energy-efficiency and performance of in-memory parallel join algorithms in database servers, we study parallelization of stream join algorithms on multicore processors to achieve energy efficiency and high performance. Equi-join is the most frequent type of join in query workloads and symmetric hash join (SHJ) algorithm is the most effective algorithm to evaluate equi-joins in data streams. To best of our knowledge, we are the first to propose a shared-memory parallel symmetric hash join algorithm on multi-core CPUs. Furthermore, we introduce a novel parallel hash-based stream join algorithm called chunk-based pairing hash join that aims at elevating data throughput and scalability. We also tackle parallel processing of multi-way stream joins where there are more than two input data streams involved in the join operation. To best of our knowledge, we are also the first to propose an in-memory parallel multi-way hash-based stream join on multicore processors. Experimental evaluation on our proposed parallel algorithms demonstrates high throughput, significant scalability, and low latency while reducing the energy consumption. Our parallel symmetric hash join and chunk-based pairing hash join achieve up to 11 times and 12.5 times more throughput, respectively, compared to that of state-of-the-art parallel stream join algorithm. Also, these two algorithms provide up to around 22 times and 24.5 times more throughput, respectively, compared to that of non-parallel (sequential) stream join computation where there is one processing thread.

The aim of this master's thesis is to design a language through which it would be possible to monitor a stream of network flows in order to detect network policy violations in the local network. An analysis of the languages used in the data stream management systems and an analysis of tasks submitted by the potential administrator were both carried out. The analysis specified resulted in the language design which represents pipelining consisting of filtering and aggregation. These operations can be clearly defined and managed within security rules. The result of this thesis also results in the Policer modul being integrated in the NEMEA system, which is able to apply the main commands of the proposed language. Finally, the module meets the requirements of the specified tasks and may be used for further development in the area of monitoring network policies.