Introduction to Scientific Computing

(winter 2001/2002)

Prof. Dr. Chr. Zenger

Prof. Dr. H.-J. Bungartz

This lecture is joint work with the lecture Introduction to Scientific Computing, given at the University of Stuttgart.

Audience:

Time and Place:

Wednesday 10:15-11:45, lecture hall 2705A

Contents:

This course provides an overview of scientific computing, i. e. of the different tasks to be tackled on the way towards powerful numerical simulations. Starting from mathematical models (derivation, analysis, and classification; various examples), their numerical treatment is discussed (discretization of differential systems, grid generation). The next chapter deals with the efficient implementation of numerical algorithms, both on monoprocessors and parallel computers (architectural features, parallel programming, load distribution, parallel numerical algorithms). Finally, some remarks on the interpretation of numerical results (visualization) are made. The course is conceived as an introduction to the thriving field of numerical simulation for computer scientists, mathematicians, engineers, or natural scientists without an already strong background in numerical methods.

Lecture Notes and other Course Material:

Lesson 1: What is Scientific Computing? PDF (4932K)
Lesson 2: Tools: Libraries and Software PDF (1404K)
Lesson 3: Principles of Mathematical Modelling PDF (1544K)
Lesson 4: Continuous Models 1: ODE PDF (952K)
Lesson 5: Continuous Models 2: PDE PDF (1000K)
Lesson 6: Numerical Treatment of ODE PDF (1116K)
Lesson 7: Numerical Treatment of ODE PDF (1240K)
Lesson 8: Standard Iterative Solvers of SLE PDF (968K)
Lesson 9: Fast Iterative Solvers of SLE PDF (692K)
Lesson 10: Implementation: Target Architectures PDF (908K)
Lesson 11: Implementation: Parallelization PDF (832K)
Lesson 12: Grid Generation and Refinement PDF (836K)
Lesson 13: Interpreting the Results: Visualization PDF (728K)
Lesson 14: Case Study: CFD PDF (660K)

Block Tutorials:

Prof. Dr. Hans-Joachim Bungartz (University of Stuttgart) and
Dr. Stefan Zimmer (University of Stuttgart)

Time and Place:
datetimeroomworksheet
Oct 26th (Friday)15:00-18:000406 tutorial.mws
Nov 16th (Friday)15:00-18:000406 diffeq.mws and worksheet including solutions
Dec 12th (Wednesday)15:00-18:000406 diffnum.mws and worksheet including solutions
Jan 18th (Friday)15:00-18:000406 linsolv.mws and the worksheet from the tutorial of Feb 1st

The four block tutorials will take place in the multimedia room (room number 0406) of the Institute for Communication Networks".

Weekly tutorials:

Dr. Stefan Zimmer (University of Stuttgart)

Time and place: Friday, 8:30-10:00, seminar room 1237

Topics:
dateTutorial
Feb 1stAfter some confusion about the tutorial last week, there will be another tutorial for questions about the worksheet on iterative methods (8:30-10:00 - if the train from Stuttgart arrives in time, some minutes later if it does not - room 1237)
Feb 8th (Friday)Deadline for submission of the exercises

Submission of exercises: please send an email to Stefan Zimmer (zimmer@in.tum.de) and attach the Maple worksheet containing your solutions to the exercises.

Final Exam:

The Final Exam in Scientific Computing will be on Thursday, February, 28th at 14:00-16:00 in room 1229.

As the midterm exam, it will consist of one part (30 minutes) with questions tat have to be answered without notes, books etc., and of a second part (90 minutes) for which notes and books are allowed, but no calculators etc.
It will cover the contents of the lectures up to and including Lesson 9 'Fast iterative solvers of SLE'.


Michael Bader