7 resultados para Escalonamento de threads

em Repositório Científico do Instituto Politécnico de Lisboa - Portugal


Relevância:

10.00% 10.00%

Publicador:

Resumo:

Neste trabalho é apresentado um Sistema de Informação (SI) cujo objetivo é contribuir para a colaboração entre VE possibilitando a sua entrada no mercado de energia elétrica, permiitindo deste modo que o proprietário do VE seja compensado. Esta compensação incidiria em medidas de minimização do investimento inicial na compra do VE e na criação de um modelo de armazenamento de energia mais sustentável. A este SI chamou-se de Broker Colaborativo ou simplesmente Broker e deverá permitir ao proprietário do VE criar um perfil com a sua informação pessoal, com informação do VE que possui e com informação relativa à colaboração que pretende efetuar, ou seja, locais e percentagem da bateria que pretende usar na colaboração. Para assegurar a entrada no mercado de energia torna-se necessário agrupar e filtrar um número significativo de VE, sendo essa responsabilidade atribuída ao Broker. Sendo a otimização do consumo de energia um fator importante neste trabalho, a implementação do Broker foi efetuada sobre o paradigma de Cloud Computing (será utilizada a designação abreviada de Cloud) onde os recursos energéticos são partilhados. A Clod possuiu uma grande capacidade de escalonamento onde se poderão criar os recursos à medida da necessidade. A utilização da Cloud permite que as especificações iniciais das máquinas virtuais, necessárias para alojar a base de dados ou os ficheiros do Broker, possam ser configuradas conforme o número de utilizadores do sistema. Assim, não se desperdiçam recursos que não seriam utilizados. Devido a estes aspetos dinâmicos, que estão intrínsecos ao fator humano, o Broker foi implementado utilizando um módulo de integração com a rede social Facebook esperando-se desta forma, que a colaboração entre diferentes VE seja difundida através dos amigos e conhecidos de cada utilizador da colaboração. Sendo assim, este trabalho pretende criar um sistema de informação onde os proprietários de VE possam juntar-se e criar uma colaboração possibilitando a sua entrada no mercado da energia elétrica. Essa colaboração será alargada recorrendo a uma rede social e compensada através de créditos. É abordado o tema das plataformas de Cloud como uma plataforma sustentável e vantajosa para o desenvolvimento de sistemas com grande potencial de crescimento.

Relevância:

10.00% 10.00%

Publicador:

Resumo:

Liquid crystals in confined geometries exhibit numerous complex structures often including topological defects that are controlled by the nematic elasticity, chirality and surface anchoring. In this work, we study the structures of cholesteric droplets pierced by cellulose fibres with planar anchoring at droplet and fibre surfaces. By varying the temperature we demonstrate the role of twisting power and droplet diameter on the equilibrium structures. The observed structures are complemented by detailed numerical simulations of possible director fields decorated by defects. Three distinct structures, a bipolar and two ring configurations, are identified experimentally and numerically. Designing cholesteric liquid crystal microdroplets on thin long threads opens new routes to produce fibre waveguides decorated with complex microresonators.

Relevância:

10.00% 10.00%

Publicador:

Resumo:

The purpose of this paper is to discuss the linear solution of equality constrained problems by using the Frontal solution method without explicit assembling. Design/methodology/approach - Re-written frontal solution method with a priori pivot and front sequence. OpenMP parallelization, nearly linear (in elimination and substitution) up to 40 threads. Constraints enforced at the local assembling stage. Findings - When compared with both standard sparse solvers and classical frontal implementations, memory requirements and code size are significantly reduced. Research limitations/implications - Large, non-linear problems with constraints typically make use of the Newton method with Lagrange multipliers. In the context of the solution of problems with large number of constraints, the matrix transformation methods (MTM) are often more cost-effective. The paper presents a complete solution, with topological ordering, for this problem. Practical implications - A complete software package in Fortran 2003 is described. Examples of clique-based problems are shown with large systems solved in core. Social implications - More realistic non-linear problems can be solved with this Frontal code at the core of the Newton method. Originality/value - Use of topological ordering of constraints. A-priori pivot and front sequences. No need for symbolic assembling. Constraints treated at the core of the Frontal solver. Use of OpenMP in the main Frontal loop, now quantified. Availability of Software.

Relevância:

10.00% 10.00%

Publicador:

Resumo:

One of the main problems of hyperspectral data analysis is the presence of mixed pixels due to the low spatial resolution of such images. Linear spectral unmixing aims at inferring pure spectral signatures and their fractions at each pixel of the scene. The huge data volumes acquired by hyperspectral sensors put stringent requirements on processing and unmixing methods. This letter proposes an efficient implementation of the method called simplex identification via split augmented Lagrangian (SISAL) which exploits the graphics processing unit (GPU) architecture at low level using Compute Unified Device Architecture. SISAL aims to identify the endmembers of a scene, i.e., is able to unmix hyperspectral data sets in which the pure pixel assumption is violated. The proposed implementation is performed in a pixel-by-pixel fashion using coalesced accesses to memory and exploiting shared memory to store temporary data. Furthermore, the kernels have been optimized to minimize the threads divergence, therefore achieving high GPU occupancy. The experimental results obtained for the simulated and real hyperspectral data sets reveal speedups up to 49 times, which demonstrates that the GPU implementation can significantly accelerate the method's execution over big data sets while maintaining the methods accuracy.

Relevância:

10.00% 10.00%

Publicador:

Resumo:

This article analyses performance consumptions among young people. The theme is explored along two main axes. The first concerns the social heterogeneity in this field, considered on two levels: the different purposes for those investments - cognitive/mental and physical performance; and the different social contexts - university and work - where performance practices and dispositions may be fostered. The second axis explores the roles of pharmacological and natural consumptions, and their interrelationship, in the dissemination of these practices. The empirical data for this analysis were drawn from an ongoing research project on performance consumptions among young people (aged 18-29 years) in Portugal, including both university students and young workers without university education. The results correspond to the stage of extensive research, for which a questionnaire was organised at a national level, using non-proportional quota sampling. On the one hand, they show that (a) there is a hierarchy of acceptance of consumptions according to their purposes, with cognitive/mental performance showing higher acceptance and (b) both pharmaceuticals and natural products are consumed for every type of performance investment. On the other, the comparison between students and workers introduces a certain heterogeneity in this general backdrop, both in terms of the purposes for their consumptions and their opting for natural or pharmacological resources. These threads of heterogeneity will prompt a discussion of the dynamics of pharmaceuticalisation within the field of performance, in particular how therapeutic cultures may be changing in terms of the way individuals relate to medications, expanding their uses in social life.