On this page:

Large-Scale and Supercomputing

"There is no problem so big it cannot be run away from."
- Charles M. Schulz

Five day intensive course (20h lectures, 10h exercises), August 15 - 19, 2011. The course is free of charge. There is 40 seats in the course, filled in registration order.

Objectives

Course gives an introduction to basic computational tools for large-scale and supercomputing in any application area. That includes basic knowledge on programming tools in Unix environment and Fortran 95 programming language; parallel programming with both the message passing interface (MPI) and OpenMP paradigms; an efficient use of CSC's metacomputer environment (e.g. data storage, batch job systems and the Scientist's Interface); the usage of Techila grid for high-performance computing; and code optimization for achieving high performance in numerical scientific computing on modern supercomputers. The course consists of lectures and hands-on exercises. At the end of the course the attendees will be able to use grid computing and write simple parallel programs as well as parallelize and optimize existing programs.

Prerequisites

The participants are assumed to have (at least low level) knowledge in some programming language. The first lectures in Fortran 95 will begin more or less from scratch. The examples are given both in Fortran 95 and C/C++.

Content

Preliminary Program

Mon, August 15: Unix Environment and Fortran 95

Tue, August 16: Techila Grid

Wed, August 17: CSC Resources and Parallel Programming

Thu, August 18: Parallel Programming

Fri, August 19: Parallel Programming and Code Optimization

Information

Registration and fees

Collaborators