Saltar al contingut Menu
Mapa
  • Inicio
  • Información
  • Contacto
  • Mapa

Programming and Tuning Massively Parallel Systems Summer School

Compartir
Introducida: 06-05-2010
This course is aimed at enriching the skills of researchers, graduate students and teachers with cutting-edge technique and hands-on experience in developing applications for many-core processors with massively parallel computing resources like GPU accelerators.
Co-Directors:  Mateo Valero (BSC/UPC) and Wen-mei Hwu (UIUC)

Instructors: Wen-mei Hwu, University of Illinois at Urbana-Champaign and David Kirk, NVIDIA Fellow, former chief science officer, NVIDIA Corporation

The summer school is organized into three main tracks according to the attendees' knowledge, expertise and interest levels. Participants should  start at the appropriate entry level, and later optionally join other tracks in order to cover more advanced topics, multiple programming languages, specific numerical methods and evaluation tools.

- Beginners Track: The Beginners Track is an introduction to CUDA, OpenCL and OpenMP with labs/examples of increasing complexity, and case studies. Attendees may then join the Advanced Track and/or Programming Models sessions.

- Teachers Track: The Teachers Track includes special sessions on programming with GPUs for current and future teachers, and labs with educational material examples. Attendees may then join the Advanced Track and/or Programming Models sessions.

- Advanced Track: The Advanced Track focuses on optimizations and tuning of scientific and engineering Applications with specific sessions for Numerical Methods and Performance Evaluation Tools, and Programming Models for Supercomputing.

- Invited Lecturers:
- BSC and UPC: Rosa Badia, Jesus Labarta, Xavier Martorell, Jose-Maria Cela, Alex Duran
- Universitat Jaume I: Enrique S. Quintana-Ortí
- Universitat Pompeu Fabra: Gianni De Fabritiis

Instructors Wen-mei Hwu and David Kirk, co-authors of "Programming Massively Parallel Processors, A Hands-on Approach", will provide students with knowledge and hands-on experience in developing applications for many-core processors, such as general purpose graphics processing units (GPUs).

Participants will have access to a multi-node cluster of GPUs, and will learn to program and optimize applications in CUDA, OpenCL, OpenMP and StaRSs. Teaching assistants will be available to help with assignments.

Case studies on Molecular Dynamics, MRI Reconstruction, RTM Stencil and Dense Linear Systems will be presented. Advanced topics will cover optimization techniques for numerical methods like FFT, Graph, Tiling, Grid, MonteCarlo, FDTD, and Sparse matrices.

- Important Dates:
Applications due: May 20
Notification of acceptance: June 1
Summer school: July 5-9, 2010

- Location: Universitat Politecnica de Catalunya UPC, Barcelona, Spain

- Complete information and program: http://bcw.ac.upc.edu

PUMPS is organized by the Barcelona Supercomputing Center, University of Illinois at Urbana-Champaign, Universitat Politecnica de Catalunya and HiPEAC Network of Excellence.

Contact: bcw2010@bcw.ac.upc.edu

Looking forward to see you in Barcelona, Mateo Valero and Wen-mei Hwu


Compartir

 
logo FIB © Facultad de Informática de Barcelona - Contacto - RSS
Esta web utiliza cookies propias para ofrecerle una mejor experiencia y servicio. Si continúa la navegación, entendemos que acepta nuestra política de cookies. Versión clássica Versión móvil