Ruprecht-Karls-Universität Heidelberg
Uni Heidelberg > IWR > MOBOCON >  Teaching > – Direct sparse LA (SS17)
[english] | [deutsch]

Seminar Direct Methods for Sparse Linear Systems

Term: Summer semester 2017
Date and Location: Wednesdays, 14-16h, SR1, Mathematikon
Kickoff meeting: Apr 19, 2017, 14-16h, room 2.414, Mathematikon
Organizer: Andreas Potschka

Contents

We consider the numerical solution of linear system of the form A x = b for the case that the system matrix A is sparse, i.e., each row of A only contains a few nonzero entries. We focus on large matrices with thousands of rows and columns. Such matrices often originate in problems on networks or after discretization of partial differential equations. In this seminar, we study algorithms that compute decompositions of A into sparse factors that allow the solution of A x = b. These methods constitute an important alternative to iterative numerical methods for sparse systems.

The seminar follows the book T.A. Davis, Direct methods for sparse linear systems, SIAM, 2006 and a few journal articles, which will be handed out at the kickoff meeting.

Target group

This seminar is targeted at MSc students of mathematics, scientific computing, and computer science.

Prerequisites

We expect knowledge equivalent to the material covered in Lineare Algebra I and Einführung in die Numerik.

Credit Points

For earning 6 CPs, the delivery of one presentation and presence at the other presentations are required.

 webmaster@iwr.uni-heidelberg.de
Last Update: 04.09.2017 - 14:48