Professor William I. Grosky
Chair of the Department of Computer and Information Science, University of Michigan-Dearborn
Tutorial Title: A Gentle Introduction to the Dimensional Reduction Landscape
In this tutorial, we compare and contrast the many linear and non-linear approaches to this technique.
Professor Yudith Cardinale
Full Professor in Computer Science Department at Universidad Simón Bolívar (USB)
Yudith Cardinale is a Full Professor in Computer Science Department at Universidad Simón Bolívar (USB) since 1996. She graduated with honors in Computer Engineering in 1990 at Universidad Centro-Occidental Lisandro Alvarado, Venezuela. She received her MSc Degree and PhD in Computer Science from USB, Venezuela, in 1993 and 2004 respectively. Her research interests include parallel processing, distributed object processing, operating systems, high performance on grid and cloud platforms, and web services composition, including web and grid semantic. She is the Director of the Parallel and Distributed Systems Group (GRyDs) at USB and coordinates several research projects. She has written a range of publications in areas such as parallel computing, grid computing, parallel checkpointing, collaborative frameworks, and Semantic Web. Her home page is http://www.ldc.usb.ve/∼yudith
Tutorial title: Ubiquitous Computing for High Performance Computing: Parallel, P2P, Grid, and Cloud Computing
Ubiquitous Computing is a big scale computing model for sharing resources with the perspective of using collaboration to enhance technology and science. The idea is to aggregate distributed resources, geographically distant, through fast networks to support high performance computing. P2P, Grid, and Cloud computing paradigm are expressions of Ubiquitous Computing, which empower an efficient and cost effective use of distributed computing resources and collaborative work among groups of users and researchers with similar scientific or commercial interests. P2P computing (also called volunteer computing) is based on the use of computers volunteered by the general public to carry out distributed scientific computing. It is a powerful way to harness distributed resources to perform large-scale tasks, similarly to other types of community-based initiatives. The main goal of the Grid Computing paradigm, targeted specifically at scientific research communities, is to transparently offer computing and data resources at the disposal of VO members. A special software layer called middleware takes care of efficient and transparent management of resources. On the other hand, the Cloud computing paradigm, mainly targeted at commercial uses, tries to fully virtualise infrastructure, platform, and applications layers, offering them as services under the so called Everything as a Service (EaaS) model. The main goal is based on the principle of economy of scale, which establishes that costs are reduced proportionally to the production volume. Parallel Computing is the main programming model to develop high performance applications. In this talk, I will present an overview of Parallel, P2P, Grid, and Cloud computing concepts, and will conduct practical exercises to introduce the parallel programming paradigm with Message Passing Interface (MPI).