PUMPS 2011: Programming and Tuning Massively Parallel Systems Summer School. (Part 11: CUDA toolkit 4.0 )
Abstract:
The PUMPS 2011 Summer School, “Programming and tUning Massively Parallel Systems (PUMPS),” was offered in July 18-22, 2011 at the Universitat Politecnica de Catalunya (UPC) campus. The event was co-sponsored by the Barcelona Supercomputing Center, UPC, the University of Illinois, HiPEAC NOE and NVIDIA, with distinguished faculty members Dr. David B. Kirk of NVIDIA and Prof. Wen-mei Hwu of the University of Illinois. PUMPS hosted 100 attendees from throughout the EU and beyond, from beginners to advanced programmers.
The content was based on CUDA, OpenCL, OpenMP, and StarSs programming languages and numerical methods including FFT, Graph, Tiling, Grid, Montecarlo, DTD, Sparse matrices, etc. The school featured lectures on cutting-edge new techniques and hands-on laboratory experience in developing applications for many-core processors with massively parallel computing resources such as the GPU accelerators, highlighting the new Fermi architecture. Applications case studies focused on Molecular Dynamics, MRI Reconstruction, RTM Stencil, and Dense Linear Systems.
More info on the summer school's website
Part 11:
This lecture covers the CUDA toolkit 4.0
David Kirk is a NVIDIA Fellow. He served from 1997 to 2009 as NVIDIA's chief scientist, a role in which he led the development of graphics technology for today’s most popular consumer entertainment platforms. He holds BS and MS degrees in mechanical engineering from the Massachusetts Institute of Technology, and MS and PhD degrees in computer science from Caltech.
| Attachment | Size |
|---|---|
| pumps_2011_barcelona_lecture9_2_cuda_toolkit_4.0_dk.pdf | 16.02 MB |
| pumps2011_lecture9_2.avi | 374.85 MB |
