|  |
02616 Large-scale Modelling |
| | |  | Danish title:
| Storskala modellering | Language:
| | Point
(ECTS )
| 5 | Course type:
|
| | |
| Schedule:
| F3A
| Scope and form: | Lectures, computer exercises and computer projects. | Duration of Course:
| 13 weeks | Date of examination:
| F3A
| Type of assessment:
| | Aid:
| | Evaluation: | | Not applicable together with: | | Qualified Prerequisites: | , | Optional Prerequisites: | |
| Participants restrictions: | Maximum: 25 | | | General course objectives:
| The advent of cost-effective multi-core computers has enabled researchers to carry out complex simulations on commodity hardware. The course aims at developing the necessary skills to exploit the potential processing power on modern computer architectures.
This course builds upon the skills aquired in the course 02614 - High-Performance Computing. It gives an overview of existing software, methods and tools for large-scale modelling.
The course enables the students to work with large-scale problems in Scientific Computing and their implementation on modern computers.
The obtained knowledge will be applied in project work to model examples of large systems. |
| Learning objectives: | | A student who has met the objectives of the course will be able to: | - analyze the performance of Scientific Computing problems
- apply modern tools for parallel programming
- design and write parallel codes for Scientific Computing problems
- apply HPC methods on large scale models
- make use of the right tools and libraries for a given problem
- use direct or iterative solution strategies
- find strategies when going from small to large problems
- apply validation techniques
- Parallel programming with MPI
| Content:
| Introduction to high-performance computing techniques, emphasizing those aspects that are important in connection with numerical computations on large data sets/large-scale problems.
Numerical methods for large-scale matrix computations, including direct and iterative methods for the solution of linear systems, the computation of eigenvalues and singular values, the solution of least-squares problems, sparse matrix computational techniques and methods for the solution of linear programs.
Application of the topics above on relevant models that require many hardware resources, e.g. memory, CPU time, etc. |
| Remarks:
| Programming and implementation is an essential part of the course. Contact the responsible course teacher about the prerequisites.
The course material consists of notes handed out by the lecturers.
The course covers modelling and the interplay between the model and methods in HPC.
The course is aimed at students from all master programs. |
| Responsible:
| , 303B, 104, (+45) 4525 3371,
, 404, 118, (+45) 4525 5603,
| Department:
| 02 Department of Informatics and Mathematical Modeling | Department involved:
| 41 Department of Mechanical Engineering | Home page:
| | Registration Sign up:
| At CampusNet | Keywords: | high-performance computers, modelling of large systems, parallel computers/algorithms, OpenMP, MPI, libraries and toolboxes |
|
|
| | Last updated:
April 27, 2012 |
See course in DTU Course base
|