Numerical Methods for Engineers I.
Code  Completion  Credits  Range  Language 

2013056  Z  3  3P+0C  Czech 
 Lecturer:
 Tutor:
 Supervisor:
 Department of Technical Mathematics
 Synopsis:

The subject deals with the numerical solution of large systems of linear or nonlinear algebraic equations, ordinary differential equations, and partial differential equations. The attention will be paid especially to the application of existing tools or frameworks for parallel computers.
 Requirements:
 Syllabus of lectures:

•Solution of linear algebra problems with dense matrices (LAPACK library)
•Parallel version of LAPACK library (ScaLAPACK)
•Direct solution of systems of linear equations with sparce matrix (UMFPACK library)
•Distribution of sparse matrices to nodes of parallel computer (spectral bisection, METIS library)
•Toolkit for numerical solution of large scale problems arising from the discretization of PDEs using parallel computers (PETSc library)
•Distributed vectors and matrices
•Iterative solvers for system of linear algebraic equations (Krylov methods)
•Solution of nonlinear algebraic equations (PETSc with SNES module)
•Numerical solution of initial value problems for ODE (PETSc with TS module)
•Numerical solution of ODE and PDE with the help of PETSc
 Syllabus of tutorials:
 Study Objective:
 Study materials:

•SAAD Yousef, Iterative Methods for Sparse Linear Systems, Second Edition, 2003,
•Lapack users'guide, http://www.netlib.org/lapack/lug/index.html,
•PETSc documentation, http://www.mcs.anl.gov/petsc/petscas/documentation/index.html
 Note:
 Further information:
 No timetable has been prepared for this course
 The course is a part of the following study plans: