Início » Tecnologia & Inovação » Webinars Técnico » 2012
04/10 - 12:00 am
Alberto Krone-Martins
Abstract ⓘ
Abstract
×Speaker
Alberto Krone-Martins
Title
Unsupervised Photometric Membership Assignment in Stellar Clusters (University of Lisboa)
Abstract
One of the most ancient problems in the photometric study of stellar clusters is the assignment of membership for its stars. Although several approaches exist for attacking this problem, they usually involve the adoption of complex theorethical models for the photometric data (isochrones) and/or the selection and use of control fields, possibly biasing some results. We have developed a data-driven, fully automated and unsupervised method to perform membership assignment in Stellar Clusters using photometric and spatial data, which is independent from complex theorethical models, as well as from the adoption of observational control fields. Our method is based on an iterative solution, and relies on Principal Component Analysis, clustering algorithms and kernel density estimations. Optionally, it also allows the user to take into account error models and missing data. We will present a description of the method, results obtained with its application to a set of realistic simulations as well as results obtained from analysis of real data of selected Open Clusters.
Slides not available
27/09 - 12:00 am
Eduardo Ogasawara
Abstract ⓘ
Abstract
×Speaker
Eduardo Ogasawara
Title
An Algebraic and Parallel Approach for Scientific Workflows using Chiron (CEFET/RJ & COPPE/UFRJ)
Abstract
Large-scale scientific experiments based on computer simulations are typically modeled as scientific workflows, which facilitate the sequencing of different programs. These scientific workflows are defined, executed and monitored by Scientific Workflow Management Systems (SWfMS). As these experiments manage large amounts of data, it is interesting, if not essential, to execute them in High Performance Computing (HPC) environments, such as clusters, grids and clouds. However, few SWfMS provide parallel support and they usually lack a run-time provenance support mechanism. Also, the existing parallel SWfMS have limited primitives to optimize workflow execution. To address these issues, we developed an algebraic approach to specify the scientific workflow, and it leaded to the implementation of Chiron: An Algebraic-Based parallel scientific workflow engine. Chiron has a native distributed provenance gathering mechanism and can perform algebraic transformations to obtain better execution strategies. Chiron is efficient in executing scientific workflows, with the benefits of bringing room for declarative specification and run-time provenance support.
Slides not available
23/08 - 12:00 am
Eduardo Bezerra
Abstract ⓘ
Abstract
×Speaker
Eduardo Bezerra
Title
Aplicações da Computação para Astronomia: Mineração de Dados e Simulação (CEFET/RJ)
Abstract
O grande volume de dados acumulado por observações em Astronomia tem suscitado a demanda por técnicas computacionais para descobrir padrões e tendências relevantes subjacentes a esses dados. De outro lado, esses mesmos dados acumulados permitem a identificação de parâmetros da dinâmica de alguns fenômenos físicos astronômicos de tal forma que esses parâmetros podem ser usados para a produção de simulações computacionais dos fenômenos. Nesse seminário, são apresentadas introduções a algumas técnicas de mineração de dados e de simulação computacional, e são apresentadas possibilidades de aplicação dessas técnicas para suporte à Astronomia.
Slides not available
16/08 - 12:00 am
Eduardo Balbinot
Abstract ⓘ
Abstract
×Speaker
Eduardo Balbinot
Title
FindSat – Hunting for sparse structures in the halo (UFRGS, LineA)
Abstract
The census of MW satellites and their distribution as a function of Galactocentric distance, mass, metallicity among other properties, will help recover the history of mass assembly in our Galaxy and test structure formation scenarios.The Milky Way has two distinct groups of satellite galaxies when we consider their year of discovery. 9 satellites were known up to the 1970s. About 15 new small galaxies were discovered with the wide and homogeneous surveys from the past 15 years, such as SDSS and 2MASS. The Dark Energy Survey (DES) is anticipated to lead to the detection of dozen more such systems, assuming that their stellar nature can be distinguished from the much larger number of galaxy clusters to be sampled. With this goal in mind we are developing FindSat, an algorithm that searches for new satellite galaxy and star clusters candidates as overdensities of stellar sources in a large photometric survey. FindSat was validated using SDSS DR7 data and successfully recovered all previously identified MW satellites with these data. It was then applied to DR8 data in the Southern Galactic region of the BOSS survey. 30 candidates were found, for which deeper follow up images from CFHT are being taken. We will present our very preliiminary results from these new images, including one already confirmed new MW satellite.
Slides not available