Browsing by Author "Alves, Albano"
Now showing 1 - 10 of 36
Results Per Page
Sort Options
- Biblioteca Digital do IPB: integração, partilha e acesso abertoPublication . Pais, Clarisse; Alves, AlbanoEm junho de 2006 foi criada a Biblioteca Digital do IPB e em novembro desse mesmo ano foram depositados os primeiros documentos. Foram dados os primeiros passos para a concretização de um objetivo plural - a criação de um repositório. É uma estrutura de apoio à investigação que promove a integração, a partilha e o acesso aberto à produção científica, realizada e produzida pela comunidade académica do Instituto Politécnico de Bragança. É, assim, um garante da preservação da memória intelectual e da investigação aplicada de referência. A implementação do repositório foi sempre apoiada pela atual presidência do IPB, facilitando, assim, a tomada de decisões e a definição da Política de Auto-Arquivo de Publicações na Biblioteca Digital do IPB. De destacar ainda que o Regulamento do Sistema de Avaliação do Desempenho do Pessoal Docente do IPB estabelece a “obrigatoriedade” de depositar as publicações científicas na Biblioteca Digital do IPB.
- Bridging the gap between cluster and grid computingPublication . Alves, Albano; Pina, AntónioThe Internet computing model with its ubiquitous networking and computing infrastructure is driving a new class of interoperable applications that benefit both from high computing power and multiple Internet connections. In this context, grids are promising computing platforms that allow to aggregate distributed resources such as workstations and clusters to solve large-scale problems. However, because most parallel programming tools were primarily developed for MPP and cluster computing, to exploit the new environment higher abstraction and cooperative interfaces are required. Rocmeμ is a platform originally designed to support the operation of multi-SAN clusters that integrates application modeling and resource allocation. In this paper we show how the underlying resource oriented computation model provides the necessary abstractions to accommodate the migration from cluster to multicluster grid enabled computing.
- clOpenCL - supporting distributed heterogeneous computing in HPC clustersPublication . Alves, Albano; Rufino, José; Pina, António; Santos, Luis PauloClusters that combine heterogeneous compute device architectures, coupled with novel programming models, have created a true alternative to traditional (homogeneous) cluster computing, allowing to leverage the performance of parallel applications. In this paper we introduce clOpenCL, a platform that supports the simple deployment and efficient running of OpenCL-based parallel applications that may span several cluster nodes, expanding the original single-node OpenCL model. clOpenCL is deployed through user level services, thus allowing OpenCL applications from different users to share the same cluster nodes and their compute devices. Data exchanges between distributed clOpenCL components rely on Open-MX, a high-performance communication library. We also present extensive experimental data and key conditions that must be addressed when exploiting clOpenCL with real applications.
- A cluster oriented model for dynamically balanced DHTsPublication . Rufino, José; Pina, António; Alves, Albano; Exposto, JoséIn this paper, we refine previous work on a model for a Distributed Hash Table (DHT) with support to dynamic balancement across a set of heterogeneous cluster nodes. We present new high-level entities, invariants and algorithms developed to increase the level of parallelism and globally reduce memory utilization. In opposition to a global distribution mechanism, that relies on complete knowledge about the current distribution of the hash table, we adopt a local approach, based on the division of the DHT into separated regions, that possess only partial knowledge of the global hash table. Simulation results confirm the hypothesis that the increasing of parallelism has as counterpart the degradation of the quality of the balancement achieved with the global approach. However, when compared with Consistent Hashing and our global approach, the same results clarify the relative merits of the extension, showing that, when properly parameterized, the model is still competitive, both in terms of the quality of the distribution and scalability.
- CoR's Faster Route over MyrinetPublication . Pina, António; Alves, Albano; Oliveira, Vitor; Moreira, CecíliaIn this paper we concentrate in the efforts made to exploit the performance of Myrinet to build a faster communication route into CoR1. By accessing the Myrinet interface through GM2, we achieved low latency and high bandwidth message passing without the overhead of a higher level protocol stack, system calls or interrupts. CoR is an ongoing project unique in its design goal of combining multithreading, message passing and distributed shared memory with facilities to dynamically select from different transport media and protocols the one that best fits communication and interaction requirements. The ability to mix CoR and PVM calls in the same program brings numerous benefits to the application developer familiar with PVM, notably: 1) new transport communication layers; PvmRouteMyrinet and PvmRouteUdp; 2) migration mechanisms for exploiting fine grain message passing; 3) thread-safe communication PVM API; 4) object-oriented distributed shared memory.
- Corporate reputation correlation as a strategic marketing tool in service design managementPublication . Maduro, Soraia; Fernandes, Paula Odete; Alves, AlbanoThis paper focused on the scope of the Design Management as a differentiating agent in the use of resources of the design and its interaction with the strategies and goals of a Service Sector, particularly in Higher Education Institutions (HEI). Currently the Reputation of HEI represents an intangible capital and a valuable asset and is recognized an important role in the differentiation and competitive advantage (Ruão, 2008). This competitive environment together with limitations of public resources for Higher Education makes the image an essential part of the strategic management of HEI (Luque-Martinez & Del Barrio Garcia, 2009). Conducting a correlations were obtained in most of the Corporate Character Scale dimensions, which reflects that these are very important when IPB's strong and sustained reputation. However, the presence of strong and positive relationships in dimensions such as ruthlessness, informality and machismo reflect the need to pay more attention to these characteristics in order to contribute to the overall satisfaction of the image perceived by the IPB Community. This paper aims to show that one of the ways to create new communication strategies can be through correlational analysis as a strategic marketing tool in Service Design Management in order to add value to the services, increasing their corporate reputation.
- A data mining approach for predicting academic success – a case studyPublication . Martins, Maria Prudência; Miguéis, Vera; Fonseca, Davide; Alves, AlbanoThe present study puts forward a regression analytic model based on the random forest algorithm, developed to predict, at an early stage, the global academic performance of the undergraduates of a polytechnic higher education institution. The study targets the universe of an institution composed of 5 schools rather than following the usual procedure of delimiting the prediction to one single specific degree course. Hence, we intend to provide the institution with one single tool capable of including the heterogeneity of the universe of students as well as educational dynamics. A different approach to feature selection is proposed, which enables to completely exclude categories of predictive variables, making the model useful for scenarios in which not all categories of data considered are collected. The introduced model can be used at a central level by the decision-makers who are entitled to design actions to mitigate academic failure.
- Deploying applications in Multi-SAN SMP clustersPublication . Alves, Albano; Pina, António; Rufino, José; Exposto, JoséThe effective exploitation of multi-SAN SMP clusters and the use of generic clusters to support complex information systems require new approaches. On the one hand, multi-SAN SMP clusters introduce another level of parallelism which is not addressed by conventional programming models that assume a homogeneous cluster. On the other hand, traditional parallel programming environments are mainly used to run scientific computations, using all available resources, and therefore applications made of multiple components, sharing cluster resources or being restricted to a particular cluster partition, are not supported. We present an approach to integrate the representation of physical resources, the modelling of applications and the mapping of application into physical resources. The abstractions we propose allow to combine shared memory, message passing and global memory paradigms.
- Deploying applications in multi-SAN SMP Clusters (extended version)Publication . Alves, Albano; Pina, António; Rufino, José; Exposto, JoséThe effective exploitation of multi-SAN SMP clusters and the use of generic clusters to support complex information systems require new approaches; multi-SAN SMP clusters introduce new levels of parallelism and traditional environments are mainly used to run scientific computations. In this paper we present a novel approach to the exploitation of clusters that allows integrating in a unique metaphor: the representation of physical resources, the modelling of applications and the mapping of application into physical resources. The proposed abstractions favoured the development of an API that allows combining and benefiting from the shared memory, message passing and global memory paradigms.
- Desconexão e replicação em aplicações distribuídasPublication . Alves, AlbanoO suporte à desconexão e a replicação são técnicas já bem conhecidas para aumentar a disponibilidade e a fiabilidade dos sistemas distribuídos. No entanto, até à data a prática comum tem sido a concretização destas ideias para casos pontuais, ou seja, cada programador implementa algoritmos deste tipo à sua maneira e de acordo com as características da aplicação em causa. Nesta dissertação são apresentados uma metodologia para dotar uma aplicação já existente, desenvolvida segundo o modelo Cliente-Servidor, de suporte à desconexão e um mecanismo para replicação de servidores. O suporte à desconexão é conseguido com a introdução de um agente entre o cliente e o servidor. O agente é obtido por um tradutor que processa a especificação da interface anotada com primitivas de desconexão. Estas primitivas constituem uma extensão à linguagem de especificação de interfaces. A replicação de servidores é obtida com a intercepção dos pedidos (invocações remotas) dos clientes e seu posterior reenvio através de um serviço de comunicação em grupo. O resultado deste trabalho é portanto o modelo Cliente-Agente-Servidores, para aplicação genérica a soluções desenvolvidas sem preocupações notórias no que respeita à disponibilidade e fiabilidade.