Mentions légales du service

Skip to content
Snippets Groups Projects
Commit 248ebf6a authored by ANDRADE-BARROSO Guillermo's avatar ANDRADE-BARROSO Guillermo
Browse files

Update README.md

parent af4989f0
No related branches found
No related tags found
No related merge requests found
# tutorials_parallel_computing # tutorials_parallel_computing
some tutorials for parallel computing courses including CPU openMP, SSE, and GPU pycuda, pyOpenCL and maybe others These tutorials are designed to work on top of '''Python'''. Python is high-level language very useful for scientific computation and rapid prototyping with '''numpy''' module. If you don't feel well with python language, don't worry, in these tutorials only the basics concepts of Python are needed and some explanations are given in important cases. If you want to know more about Python language please visit [http://python.org/ Official Python web site] or read [https://developers.google.com/edu/python/ Google Python courses]
\ No newline at end of file
Work with these tutorials supposes that you have a python interpreter installed with numpy and scipy installed as well as g++ compiler. For Cuda and OpenCL you have to install '''PyCuda''' and '''PyOpenCL''' modules. But INRIA Rennes offers to you the possibility to get access to an experimental server (named ''soroban'') with all you need. Please feel free to request to [mailto:Guillermo.Andrade-Barroso@inria.fr?subject=Request%20for%20acaccount%20in%20soroban%20server&body=Hello%2C%0AI%20want%20to%20access%20to%20soroban%20server%0AYou%20can%20find%20my%20ssh%20public%20key%20attached%20to%20this%20message.%0AMy%20INRIA%20team%20is:.......%0AI%20need%20access%20for%20about%20XXX%20Days/Months%20to%20Work%20on%20this%20topics:%0A...............%0A...............%0ABest%20Regards%2C Guillermo Andrade B.] an account to this machine with ssh connection.
[[Media:Introduction_au_Calcul_Parallele2_EN.pdf|'''Parallel Computing Courses Slides here''']]
== SSE and OpenMP ==
=== Introduction to C++ SSE using Python and Weave ===
There is a tutorial showing :
* how to include C++ code in python script with scipy.weave
* how to optimize C++ code using SSE and measure performance
[[SSE tutorial using Python and Weave module]]
=== Multi-core programming with OpenMP and SSE using Python and Weave ===
There is a tutorial showing :
* how to optimize C++ code using OpenMP and SSE
[[OpenMP and SSE tutorial using Python and Weave module]]
== CUDA ==
=== Blocks and Grid on scalar multiplication kernel===
The aim of this first tutorial is to learn how to define grids and blocks to work with a linear arbitrary size algebra problem (scalar multiplication) :
[[CUDA Grid and Blocks tutorial]]
=== Reverse kernel and performance===
The goal of this tutorial is to understand how pipeline continuity in memory access may improve performance is some algorithms using GPUs. For this tutorial will work with a naive reverse memory algorithm.<br />
[[Pipeline performance in reverse kernel|Pipeline performance in reverse kernel]]
== OpenCL ==
=== Scan Amino Acid sequence ===
Naive scan of Animo Acid sequences for seeking fragment (protein) inside.<br />
[[Scan Amino Acid sequence]]
=== Brute force for solving a puzzle ===
MAN at WORK ...
Basic method for scan configurations of parts placement in a puzzle using OpenCL and comparison of architectures
[[Brute force for solving a puzzle]]
[[category: Documentation ]]
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment