Vicente Blanco

Publications (80) View all

• Source

  Available from: Vicente Blanco

  Article: Memoria TAD: Experiencias con Python y CUDA en Computación de Altas Prestaciones

  Sergio Armas, Lionel Mena, Alejandro Samarín, Vicente Blanco, Alberto Morales, Francisco Almeida
  [show abstract] [hide abstract]
  ABSTRACT: El aprovechamiento de la capacidad de cómputo de los dispositivos gráficos para resolver problemas computacionalmente complejos está en auge. El alto grado de paralelismo que dichos dispositivos proveen, además de la disponibilidad de kits especializados de desarrollo de software para el público general, abren la puerta a nuevas formas de resolver problemas científicos en una fracción del tiempo que emplearían algoritmos similares basados en CPU. El siguiente paso es encontrar el equilibrio entre la potencia de estos paradigmas de programación y la flexibilidad de los lenguajes modernos. Es aquí donde PyCUDA [1] entra en escena; un "wrapper" de la programación CUDA para Python que ofrece al programador el acceso a la computación de altas prestaciones sobre dispositivos gráficos sin abandonar la comodidad y el dinamismo de este lenguaje (orientación a objetos, tipado dinámico, intérprete interactivo, etc.). Nuestros objetivos se centran, por un lado, en preparar una máquina de prueba equipada con el hardware necesario y, por otro, comprobar las facilidades que promete PyCUDA así como su rendimiento frente a problemas reales.
  08/2011;
• Source

  Available from: E M Garzon

  Article: Adaptive load balancing of iterative computation on heterogeneous nondedicated systems.

  J. A. Martínez, F. Almeida, E. M. Garzón, A. Acosta, V. Blanco
  The Journal of Supercomputing. 01/2011; 58:385-393.
• Conference Proceeding: Dynamic load balancing on heterogeneous multicore/multiGPU systems

  Alejandro Acosta, Robert Corujo, Vicente Blanco, Francisco Almeida
  [show abstract] [hide abstract]
  ABSTRACT: Parallel computing in heterogeneous environments is drawing considerable attention due to the growing number of these kind of systems. Adapting existing code and libraries to such systems is a fundamental problem. The performance of this code is affected by the large interdependence between the code and these parallel architectures. We have developed a dynamic load balancing library that allows parallel code to be adapted to heterogeneous systems for a wide variety of problems. The overhead introduced by our system is minimal and the cost to the programmer negligible. The strategy was applied to a Dynamic Programming Algorithm, the Resource Allocation Problem. This code has been implemented on different heterogeneous architectures, including an heterogeneous cluster, a multicore system, a single GPU, and a multi-GPU system. The unbalance nature of the RAP algorithm shows the success of our load balancing library on such architectures.
  High Performance Computing and Simulation (HPCS), 2010 International Conference on; 08/2010
• Conference Proceeding: El criterio de información de Akaike en la obtención de modelos estadísticos de Rendimiento

  Diego R Martínez, Julio L Albín, José C Cabaleiro, Tomás F Pena, Francisco F Rivera, Vicente Blanco
  XX Jornadas de Paralelismo; 09/2009
• Article: Using Web Services for Performance Monitoring and Scheduling

  Adrian Santos, Francisco Almeida, Vicente Blanco, David Diez, Jonay Regueira, Esau Sicilia
  [show abstract] [hide abstract]
  ABSTRACT: The adoption of Web Service standards provides us with an increased level of manageability, extensibility and interoperability between loosely coupled services.The adoption of Web Services technologies atop sites for performance monitoring and scheduling will improve the efficient use of the computational resources. Web Services provide the ability to decompose HPC resources and functionality into a set of discoverable and loosely coupled services, which are capable of interaction in heterogeneous environments. At the same time, Web Services can address many of the interoperability issues that can be encountered in large scale systems.End users can access to these services to decide which system will be the most suitable for their needs. Other tools like schedulers can use the resources available as services to optimize the HPC resources and minimize jobs waiting time.
  16th Euromicro Conference on Parallel, Distributed and Network-Based Processing (PDP 2008). 02/2009;