Parallel Numerics
(winter 2004/05)

Prof. Dr. Th. Huckle

Audience:

• students in Computational Science and Engineering (CSE, compulsory course)
• students in Mathematik/Technomathematik (elective)

Time and Place:

Results of the written examination:

Contents:

• High-Performance Computing,
• Performance: Analysis, Modeling, and Measurements,
• Basic Linear Algebra Subprograms,
• Direct Solution of Sparse Linear Systems,
• Iterative Methods for Linear Systems,
• Linear Eigenvalue Problems,
• ...

Lecture Notes and other Course Material:

Lecture Notes (last year's lecture, PDF document)

Exercise 1	: MPI (recapitulation)	blatt01.c machines.dat blatt01.ps
Exercise 2	: Numerical Integration	Solution 1: serial program blatt02.ps blatt02_2.c
Exercise 3	: Vector-Vector Operations	Solution 1: basic program Solution 2: program using collective communication routines blatt03.ps
Exercise 4	: Matrix-Matrix Multiplication	Solution 1: serial program Solution 2: basic parallel program blatt04.ps
Exercise 5	: Parallel Gauss Elimination	blatt05.ps blatt05_2.c blatt05_3.c
Exercise 6	: Sparse Matrix Vector	blatt06.c
Exercise 7	: Partitioning Method	blatt07.c
Exercise 8	: SOR	blatt08.c blatt08_2.c blatt08_3.c
Exercise 9	: Conjugate Gradient Method	blatt09.c

Tutorial: