SEMINARIS D´EMPRESA

Sessió 1  (8 de febrer)
11.00 a 14.00 Earth Science Department BSC
Convener: Kim Serradell

A series of talks will introduce the research done in the Earth Sciences Department of the Barcelona Supercomputing Center on Climate Prediction, Atmospheric Composition and Computational Earth Sciences as well as the Earth System Services derived from the activities of the Department.

 The Department develops and applies state-of-the-art Earth Systems Models to study the evolution of the climate, the atmospheric composition and their interrelations.

Agenda:
Climate prediction/ES Services (45 mins)
Atmospheric Composition/ES Services (45 mins)
Break (15 mins)
Computational Earth Sciences (45 mins)
Discussion (30 mins)

Sessió 2 (9 de febrer)
11:00 a 12:30, Xavier Martorell: Programming Models.
12:30 a 14.00 Javier Espinosa: Visualization for data centric sciences in the era of Big Data

PROGRAMMING MODELS

This session will introduce the group of Programming Models @BSC, including the main lines of research on simplifying the programming of current and future heterogenous architectures. We will show the development of the OmpSs programming model, and how we support SMP multicore machines, GPUs, and FPGAs.

Agenda:

- Programming Models group @ BSC (15')

- The OmpSs Programming Model (45')

- Hands-on demo on various applications and environments (30')

Visualization for data centric sciences in the era of Big Data

The beginning of the XXI century is characterized by a rapid and massive production of data called Big Data, which has led to the emergence of “new” data centric sciences (data science, data journalism, network science, digital humanities, digital urbanism) and the definition of new expertise profiles for dealing with it (data scientist/engineer/curator). This has led to new requirements and challenges. In this new context, data visualization stands as a simple yet powerful and interdisciplinary tool for doing findings and/or confirming hypothesis in data centric sciences. This lecture will address different techniques for visualizing Big Data considering Big Data visualization as a complex and greedy task composed of several steps (data collection, cleansing, integration, analysis and visualization).


Sessió 3 (10 de febrero)
11 a 12:30 Mariano Vázquez: Supercomputación en investigación biomédica: sistema cardiovascular

Descripción: En la charla se mostrarán ejemplos de aplicación de la supercomputación en simulaciones a nivel órgano del sistema cardiovascular. Las simulaciones están basadas en el código Alya, desarrollado en el BSC. Se mostrarán simulaciones sobre el corazón y su función, de despliegue de stents en arterias o de uso de biochips computacionales en industria farmacéutica.


12:30 a 14.00 Issac Rudomin, Hugo Pérez, Jorge Ramírez Flores, SIMULATING AND VISUALIZING REAL-TIME CROWDS ON GPU CLUSTERS

Abstract: We present a set of algorithms for simulating and visualizing real-time crowds in GPU (Graphics Processing Units) clusters. First we present crowd simulation and rendering techniques that take advantage of single GPU machines. Then, using as an example a wandering crowd behavior simulation algorithm, we explain how this kind of algorithms
can be extended for their use in GPU cluster environments. We also present a visualization architecture that renders the simulation results using detailed 3D virtual characters. This architecture is adaptable in order to support the Barcelona Supercomputing Center
(BSC) infrastructure. The results show that our algorithms are scalable in different hardware platforms including embedded systems, desktop GPUs, and GPU clusters, in particular, the BSCs Minotauro cluster.