Computer architectures and optimisation

In this training course, the architectures of today's supercomputers are described. The processor evolution as well as the main types of architectures are then presented. Finally, the methods and tools for code optimisation on these supercomputers are described.

Targent participants:
This training session is for students, engineers, physicists and
computer scientists wishing to discover today's supercomputer
architectures and to learn methods and tools for code optimisation
on these supercomputers.

Prerequisites:
None.

Scientific contact:
Isabelle d'AST

Fee:
- Public: 300 €
- CERFACS shareholders/CNRS/INRIA: 150 €
- Students: 50 €

Dates: 29 March 2013

Deadline for registration: 15 days before the starting date of the training course

Duration: 1 day
(5 hours eligible for Individual Right to Training (DIF))

Program

(From 9h30 to 16h30)

Morning: Introduction to supercomputer architectures

• Introduction

• Processor evolution

• Main types of architecture

• Examples of today's common supercomputers

Afternoon: Methods and tools for code optimisation

• Sequential and parallel optimisations

• Methodology of code optimisation process

• Examples of software tools for code optimisation

Back to main program