Multicore Scheduling of Real-Time Irregular Parallel Algorithms in Linux

Face à estagnação da tecnologia uniprocessador registada na passada década, aos principais fabricantes de microprocessadores encontraram na tecnologia multi-core a resposta `as crescentes necessidades de processamento do mercado. Durante anos, os desenvolvedores de software viram as suas aplicações acompanhar os ganhos de performance conferidos por cada nova geração de processadores sequenciais, mas `a medida que a capacidade de processamento escala em função do número de processadores, a computação sequencial tem de ser decomposta em várias partes concorrentes que possam executar em paralelo, para que possam utilizar as unidades de processamento adicionais e completar mais rapidamente. A programação paralela implica um paradigma completamente distinto da programação sequencial. Ao contrário dos computadores sequenciais tipificados no modelo de Von Neumann, a heterogeneidade de arquiteturas paralelas requer modelos de programação paralela que abstraiam os programadores dos detalhes da arquitectura e simplifiquem o desenvolvimento de aplicações concorrentes. Os modelos de programação paralela mais populares incitam os programadores a identificar instruções concorrentes na sua lógica de programação, e a especificá-las sob a forma de tarefas que possam ser atribuídas a processadores distintos para executarem em simultâneo. Estas tarefas são tipicamente lançadas durante a execução, e atribuídas aos processadores pelo motor de execução subjacente. Como os requisitos de processamento costumam ser variáveis, e não são conhecidos a priori, o mapeamento de tarefas para processadores tem de ser determinado dinamicamente, em resposta a alterações imprevisíveis dos requisitos de execução. `A medida que o volume da computação cresce, torna-se cada vez menos viável garantir as suas restrições temporais em plataformas uniprocessador. Enquanto os sistemas de tempo real se começam a adaptar ao paradigma de computação paralela, há uma crescente aposta em integrar execuções de tempo real com aplicações interativas no mesmo hardware, num mundo em que a tecnologia se torna cada vez mais pequena, leve, ubíqua, e portável. Esta integração requer soluções de escalonamento que simultaneamente garantam os requisitos temporais das tarefas de tempo real e mantenham um nível aceitável de QoS para as restantes execuções. Para tal, torna-se imperativo que as aplicações de tempo real paralelizem, de forma a minimizar os seus tempos de resposta e maximizar a utilização dos recursos de processamento. Isto introduz uma nova dimensão ao problema do escalonamento, que tem de responder de forma correcta a novos requisitos de execução imprevisíveis e rapidamente conjeturar o mapeamento de tarefas que melhor beneficie os critérios de performance do sistema. A técnica de escalonamento baseado em servidores permite reservar uma fração da capacidade de processamento para a execução de tarefas de tempo real, e assegurar que os efeitos de latência na sua execução não afectam as reservas estipuladas para outras execuções. No caso de tarefas escalonadas pelo tempo de execução máximo, ou tarefas com tempos de execução variáveis, torna-se provável que a largura de banda estipulada não seja consumida por completo. Para melhorar a utilização do sistema, os algoritmos de partilha de largura de banda (capacity-sharing) doam a capacidade não utilizada para a execução de outras tarefas, mantendo as garantias de isolamento entre servidores. Com eficiência comprovada em termos de espaço, tempo, e comunicação, o mecanismo de work-stealing tem vindo a ganhar popularidade como metodologia para o escalonamento de tarefas com paralelismo dinâmico e irregular. O algoritmo p-CSWS combina escalonamento baseado em servidores com capacity-sharing e work-stealing para cobrir as necessidades de escalonamento dos sistemas abertos de tempo real. Enquanto o escalonamento em servidores permite partilhar os recursos de processamento sem interferências a nível dos atrasos, uma nova política de work-stealing que opera sobre o mecanismo de capacity-sharing aplica uma exploração de paralelismo que melhora os tempos de resposta das aplicações e melhora a utilização do sistema. Esta tese propõe uma implementação do algoritmo p-CSWS para o Linux. Em concordância com a estrutura modular do escalonador do Linux, ´e definida uma nova classe de escalonamento que visa avaliar a aplicabilidade da heurística p-CSWS em circunstâncias reais. Ultrapassados os obstáculos intrínsecos `a programação da kernel do Linux, os extensos testes experimentais provam que o p-CSWS ´e mais do que um conceito teórico atrativo, e que a exploração heurística de paralelismo proposta pelo algoritmo beneficia os tempos de resposta das aplicações de tempo real, bem como a performance e eficiência da plataforma multiprocessador.

What is the value of value-at-risk after all?: A conditional approach using quantile regressions

A Work Project, presented as part of the requirements for the Award of a Masters Degree in Finance from the NOVA – School of Business and Economics

A flexible navigation system for autonomous robots operating in heterogeneous environments

Dissertação para obtenção do Grau de Mestre em Engenharia Electrotécnica e de Computadores

Data queries over heterogeneous sources

Dissertação para obtenção do Grau de Mestre em Engenharia Informática

Time series morphological analysis applied to biomedical signals events detection

Dissertation submitted in the fufillment of the requirements for the Degree of Master in Biomedical Engineering

Multi-GPU support on the marrow algorithmic skeleton framework

Dissertação para obtenção do Grau de Mestre em Engenharia Informática

The Edges of Areal Units: a case study in the heterogeneous effects of assessment district edges

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies

Petri net based development of globally-asynchronous locally-synchronous distributed embedded systems

Dissertação para obtenção do Grau de Doutor em Engenharia Electrotécnica e de Computadores

Heterogeneous computing with an algorithmic skeleton framework

The Graphics Processing Unit (GPU) is present in almost every modern day personal computer. Despite its specific purpose design, they have been increasingly used for general computations with very good results. Hence, there is a growing effort from the community to seamlessly integrate this kind of devices in everyday computing. However, to fully exploit the potential of a system comprising GPUs and CPUs, these devices should be presented to the programmer as a single platform. The efficient combination of the power of CPU and GPU devices is highly dependent on each device’s characteristics, resulting in platform specific applications that cannot be ported to different systems. Also, the most efficient work balance among devices is highly dependable on the computations to be performed and respective data sizes. In this work, we propose a solution for heterogeneous environments based on the abstraction level provided by algorithmic skeletons. Our goal is to take full advantage of the power of all CPU and GPU devices present in a system, without the need for different kernel implementations nor explicit work-distribution.To that end, we extended Marrow, an algorithmic skeleton framework for multi-GPUs, to support CPU computations and efficiently balance the work-load between devices. Our approach is based on an offline training execution that identifies the ideal work balance and platform configurations for a given application and input data size. The evaluation of this work shows that the combination of CPU and GPU devices can significantly boost the performance of our benchmarks in the tested environments, when compared to GPU-only executions.

Impact of using different radiation therapy techniques in breat cancer: contralateral breast dose

RESUMO: O cancro de mama e o mais frequente diagnoticado a indiv duos do sexo feminino. O conhecimento cientifico e a tecnologia tem permitido a cria ção de muitas e diferentes estrat egias para tratar esta patologia. A Radioterapia (RT) est a entre as diretrizes atuais para a maioria dos tratamentos de cancro de mama. No entanto, a radia ção e como uma arma de dois canos: apesar de tratar, pode ser indutora de neoplasias secund arias. A mama contralateral (CLB) e um orgão susceptivel de absorver doses com o tratamento da outra mama, potenciando o risco de desenvolver um tumor secund ario. Nos departamentos de radioterapia tem sido implementadas novas tecnicas relacionadas com a radia ção, com complexas estrat egias de administra ção da dose e resultados promissores. No entanto, algumas questões precisam de ser devidamente colocadas, tais como: E seguro avançar para tecnicas complexas para obter melhores indices de conformidade nos volumes alvo, em radioterapia de mama? O que acontece aos volumes alvo e aos tecidos saudaveis adjacentes? Quão exata e a administração de dose? Quais são as limitações e vantagens das técnicas e algoritmos atualmente usados? A resposta a estas questões e conseguida recorrendo a m etodos de Monte Carlo para modelar com precisão os diferentes componentes do equipamento produtor de radia ção(alvos, ltros, colimadores, etc), a m de obter uma descri cão apropriada dos campos de radia cão usados, bem como uma representa ção geometrica detalhada e a composição dos materiais que constituem os orgãos e os tecidos envolvidos. Este trabalho visa investigar o impacto de tratar cancro de mama esquerda usando diferentes tecnicas de radioterapia f-IMRT (intensidade modulada por planeamento direto), IMRT por planeamento inverso (IMRT2, usando 2 feixes; IMRT5, com 5 feixes) e DCART (arco conformacional dinamico) e os seus impactos em irradia ção da mama e na irradia ção indesejada dos tecidos saud aveis adjacentes. Dois algoritmos do sistema de planeamento iPlan da BrainLAB foram usados: Pencil Beam Convolution (PBC) e Monte Carlo comercial iMC. Foi ainda usado um modelo de Monte Carlo criado para o acelerador usado (Trilogy da VARIAN Medical Systems), no c odigo EGSnrc MC, para determinar as doses depositadas na mama contralateral. Para atingir este objetivo foi necess ario modelar o novo colimador multi-laminas High- De nition que nunca antes havia sido simulado. O modelo desenvolvido est a agora disponí vel no pacote do c odigo EGSnrc MC do National Research Council Canada (NRC). O acelerador simulado foi validado com medidas realizadas em agua e posteriormente com c alculos realizados no sistema de planeamento (TPS).As distribui ções de dose no volume alvo (PTV) e a dose nos orgãos de risco (OAR) foram comparadas atrav es da an alise de histogramas de dose-volume; an alise estati stica complementar foi realizadas usando o software IBM SPSS v20. Para o algoritmo PBC, todas as tecnicas proporcionaram uma cobertura adequada do PTV. No entanto, foram encontradas diferen cas estatisticamente significativas entre as t ecnicas, no PTV, nos OAR e ainda no padrão da distribui ção de dose pelos tecidos sãos. IMRT5 e DCART contribuem para maior dispersão de doses baixas pelos tecidos normais, mama direita, pulmão direito, cora cão e at e pelo pulmão esquerdo, quando comparados com as tecnicas tangenciais (f-IMRT e IMRT2). No entanto, os planos de IMRT5 melhoram a distribuição de dose no PTV apresentando melhor conformidade e homogeneidade no volume alvo e percentagens de dose mais baixas nos orgãos do mesmo lado. A t ecnica de DCART não apresenta vantagens comparativamente com as restantes t ecnicas investigadas. Foram tamb em identi cadas diferen cas entre os algoritmos de c alculos: em geral, o PBC estimou doses mais elevadas para o PTV, pulmão esquerdo e cora ção, do que os algoritmos de MC. Os algoritmos de MC, entre si, apresentaram resultados semelhantes (com dferen cas at e 2%). Considera-se que o PBC não e preciso na determina ção de dose em meios homog eneos e na região de build-up. Nesse sentido, atualmente na cl nica, a equipa da F sica realiza medi ções para adquirir dados para outro algoritmo de c alculo. Apesar de melhor homogeneidade e conformidade no PTV considera-se que h a um aumento de risco de cancro na mama contralateral quando se utilizam t ecnicas não-tangenciais. Os resultados globais dos estudos apresentados confirmam o excelente poder de previsão com precisão na determinação e c alculo das distribui ções de dose nos orgãos e tecidos das tecnicas de simulação de Monte Carlo usados.---------ABSTRACT:Breast cancer is the most frequent in women. Scienti c knowledge and technology have created many and di erent strategies to treat this pathology. Radiotherapy (RT) is in the actual standard guidelines for most of breast cancer treatments. However, radiation is a two-sword weapon: although it may heal cancer, it may also induce secondary cancer. The contralateral breast (CLB) is a susceptible organ to absorb doses with the treatment of the other breast, being at signi cant risk to develop a secondary tumor. New radiation related techniques, with more complex delivery strategies and promising results are being implemented and used in radiotherapy departments. However some questions have to be properly addressed, such as: Is it safe to move to complex techniques to achieve better conformation in the target volumes, in breast radiotherapy? What happens to the target volumes and surrounding healthy tissues? How accurate is dose delivery? What are the shortcomings and limitations of currently used treatment planning systems (TPS)? The answers to these questions largely rely in the use of Monte Carlo (MC) simulations using state-of-the-art computer programs to accurately model the di erent components of the equipment (target, lters, collimators, etc.) and obtain an adequate description of the radiation elds used, as well as the detailed geometric representation and material composition of organs and tissues. This work aims at investigating the impact of treating left breast cancer using di erent radiation therapy (RT) techniques f-IMRT (forwardly-planned intensity-modulated), inversely-planned IMRT (IMRT2, using 2 beams; IMRT5, using 5 beams) and dynamic conformal arc (DCART) RT and their e ects on the whole-breast irradiation and in the undesirable irradiation of the surrounding healthy tissues. Two algorithms of iPlan BrainLAB TPS were used: Pencil Beam Convolution (PBC)and commercial Monte Carlo (iMC). Furthermore, an accurate Monte Carlo (MC) model of the linear accelerator used (a Trilogy R VARIANR) was done with the EGSnrc MC code, to accurately determine the doses that reach the CLB. For this purpose it was necessary to model the new High De nition multileaf collimator that had never before been simulated. The model developed was then included on the EGSnrc MC package of National Research Council Canada (NRC). The linac was benchmarked with water measurements and later on validated against the TPS calculations. The dose distributions in the planning target volume (PTV) and the dose to the organs at risk (OAR) were compared analyzing dose-volume histograms; further statistical analysis was performed using IBM SPSS v20 software. For PBC, all the techniques provided adequate coverage of the PTV. However, statistically significant dose di erences were observed between the techniques, in the PTV, OAR and also in the pattern of dose distribution spreading into normal tissues. IMRT5 and DCART spread low doses into greater volumes of normal tissue, right breast, right lung, heart and even the left lung than tangential techniques (f-IMRT and IMRT2). However,IMRT5 plans improved distributions for the PTV, exhibiting better conformity and homogeneity in target and reduced high dose percentages in ipsilateral OAR. DCART did not present advantages over any of the techniques investigated. Di erences were also found comparing the calculation algorithms: PBC estimated higher doses for the PTV, ipsilateral lung and heart than the MC algorithms predicted. The MC algorithms presented similar results (within 2% di erences). The PBC algorithm was considered not accurate in determining the dose in heterogeneous media and in build-up regions. Therefore, a major e ort is being done at the clinic to acquire data to move from PBC to another calculation algorithm. Despite better PTV homogeneity and conformity there is an increased risk of CLB cancer development, when using non-tangential techniques. The overall results of the studies performed con rm the outstanding predictive power and accuracy in the assessment and calculation of dose distributions in organs and tissues rendered possible by the utilization and implementation of MC simulation techniques in RT TPS.

A public transport bus assignment problem: parallel metaheuristics assessment

Combinatorial Optimization Problems occur in a wide variety of contexts and generally are NP-hard problems. At a corporate level solving this problems is of great importance since they contribute to the optimization of operational costs. In this thesis we propose to solve the Public Transport Bus Assignment problem considering an heterogeneous fleet and line exchanges, a variant of the Multi-Depot Vehicle Scheduling Problem in which additional constraints are enforced to model a real life scenario. The number of constraints involved and the large number of variables makes impracticable solving to optimality using complete search techniques. Therefore, we explore metaheuristics, that sacrifice optimality to produce solutions in feasible time. More concretely, we focus on the development of algorithms based on a sophisticated metaheuristic, Ant-Colony Optimization (ACO), which is based on a stochastic learning mechanism. For complex problems with a considerable number of constraints, sophisticated metaheuristics may fail to produce quality solutions in a reasonable amount of time. Thus, we developed parallel shared-memory (SM) synchronous ACO algorithms, however, synchronism originates the straggler problem. Therefore, we proposed three SM asynchronous algorithms that break the original algorithm semantics and differ on the degree of concurrency allowed while manipulating the learned information. Our results show that our sequential ACO algorithms produced better solutions than a Restarts metaheuristic, the ACO algorithms were able to learn and better solutions were achieved by increasing the amount of cooperation (number of search agents). Regarding parallel algorithms, our asynchronous ACO algorithms outperformed synchronous ones in terms of speedup and solution quality, achieving speedups of 17.6x. The cooperation scheme imposed by asynchronism also achieved a better learning rate than the original one.

Analysis of classification algorithms for crop detection using LANDSAT 8 images

Remote sensing - the acquisition of information about an object or phenomenon without making physical contact with the object - is applied in a multitude of different areas, ranging from agriculture, forestry, cartography, hydrology, geology, meteorology, aerial traffic control, among many others. Regarding agriculture, an example of application of this information is regarding crop detection, to monitor existing crops easily and help in the region’s strategic planning. In any of these areas, there is always an ongoing search for better methods that allow us to obtain better results. For over forty years, the Landsat program has utilized satellites to collect spectral information from Earth’s surface, creating a historical archive unmatched in quality, detail, coverage, and length. The most recent one was launched on February 11, 2013, having a number of improvements regarding its predecessors. This project aims to compare classification methods in Portugal’s Ribatejo region, specifically regarding crop detection. The state of the art algorithms will be used in this region and their performance will be analyzed.

Genetic algorithms for the non emergency patients transport service

The authors would like to thank the anonymous reviewers for their valuable comments and suggestions to improve the paper. The authors would like to thank Dr. Elaine DeBock for reviewing the manuscript.

Comparison of single and multi-objective evolutionary algorithms for robust link-state routing

Traffic Engineering (TE) approaches are increasingly impor- tant in network management to allow an optimized configuration and resource allocation. In link-state routing, the task of setting appropriate weights to the links is both an important and a challenging optimization task. A number of different approaches has been put forward towards this aim, including the successful use of Evolutionary Algorithms (EAs). In this context, this work addresses the evaluation of three distinct EAs, a single and two multi-objective EAs, in two tasks related to weight setting optimization towards optimal intra-domain routing, knowing the network topology and aggregated traffic demands and seeking to mini- mize network congestion. In both tasks, the optimization considers sce- narios where there is a dynamic alteration in the state of the system, in the first considering changes in the traffic demand matrices and in the latter considering the possibility of link failures. The methods will, thus, need to simultaneously optimize for both conditions, the normal and the altered one, following a preventive TE approach towards robust configurations. Since this can be formulated as a bi-objective function, the use of multi-objective EAs, such as SPEA2 and NSGA-II, came nat- urally, being those compared to a single-objective EA. The results show a remarkable behavior of NSGA-II in all proposed tasks scaling well for harder instances, and thus presenting itself as the most promising option for TE in these scenarios.

An immunity based configuration for multilayer single featured biometric authentication algorithms

Immune systems have been used in the last years to inspire approaches for several computational problems. This paper focus on behavioural biometric authentication algorithms’ accuracy enhancement by using them more than once and with different thresholds in order to first simulate the protection provided by the skin and then look for known outside entities, like lymphocytes do. The paper describes the principles that support the application of this approach to Keystroke Dynamics, an authentication biometric technology that decides on the legitimacy of a user based on his typing pattern captured on he enters the username and/or the password and, as a proof of concept, the accuracy levels of one keystroke dynamics algorithm when applied to five legitimate users of a system both in the traditional and in the immune inspired approaches are calculated and the obtained results are compared.