Course Materials by Subject of CPS-615,
Introduction to Computational Science, Fall 1997
Synopsis of the course
This course is the graduate level introductory course in the
discipline of Computational Science (the computer simulation of
natural systems). This course is designed to teach the basic tools
from mathematics and computer science that are needed to give
computational solutions to scientific and engineering problems.
This is a new version of the CPS615 course materials, replacing those
from Fall 1996. The course is being taught by Professor Geoffrey Fox in Fall
1997, and the materials will be updated during the semester.
Introduction to High Performance Computing (HPC) and Compuational Science
Parallel Computing and Architectures
Algorithms and Applications
- Lecture slides: Statistics and
Random Numbers, in preparation for Monte Carlo Integrals.
- Background materials on
Random Number Generators (RNG), from the
CSEP book.
- Lecture slides: ODE's and Computing
Techniques for Particle Systems, and the same material appeared in
written text.
- Background materials on
Ordinary Differential Equations, from the
CSEP book.
- Numerical Integration Module
including details on Monte Carlo Methods.
- Lecture slides: Adaptive
Parallel Computing Techniques for Numerical Integration apply parallel computing to
numerical integration.
- Lecture slides:
Iterative Methods for Partial Differential Equations.
- Lecture slides: Motivation
for Partial Differential Equations where this has extensive discussion of CFD and the
Navier Stokes equation which was prepared for CPS713 and can be ignored here. The initial
remarks on relative role of PDE's, Monte Carlo, and particle dynamics has value.
- Further references are:
Basic Discussion of PDE's in CSEP Project.
- More examples in
Material Balance Section.
- A very good discussion of iterative methods is in
Templates for the Solution
of Linear Systems: Building Blocks for Iterative Methods, by
Dongarra et al.
- A good introduction to numerical linear algebra comes from the
HPSC group at
the University of Colorado covering basic matrix analysis, elementary ODE's and Fourier
Analysis.
- The full HPSC material is available in postscript by ftp. You may need to fetch
the numerical analysis module for use in the course.
- Lecture slides: Linear
Algebra Presentation that continues from the HPSC resource with eigenvalues and
eigenvectors.
- Lecture slides: Finite
Element and Conjugate Gradient Presentation with analogies to minimization methods.
- Lecture slides: Parallel
Full Matrix Algorithms.
- More parallel computing materials, algorithms, tutorials are available at
Computational Science
page.
- On mesh and grid generation take a look at Finite Element Mesh Generation.
- For a good survey on N-Body Methods Resources, see
this page at EPCC.
Computer Architecture
- Architecture Classes of High Performance Computers, with
a discussion of:
- This starts by considering the analytic form
for communication overhead and illustrates its
stencil dependence in simple local cases --
stressing relevance of grain size. This was
covered in Homework 3.
- The implication for scaling and generalizing
from Laplace example is covered. We covered
scaled speedup (fixed grain size) as well as
fixed problem size. Also noted some useful
material was missing and this was continued in
next lecture (Sept 10) .
- The lecture starts coverage of computer
architecture covering base technologies with
both CMOS covered in an earlier lecture
contrasted to Quantum and Superconducting technology.
- Audio Recording: on performance and computer technologies.
Lecture (3) foilsets -- September 10, 1996
- Comments: start with foil
MIMD Distributed Memory Architecture, with a discussion of:
- Details of communication overhead in parallel
processing for case where "range" of
interaction is large.
- Two old examples from Caltech illustrating
correctness of analytic form.
- Return to more discussion on computer
architectures describing:
- Vector Supercomputers.
- General Relevance of data locality and pipelining.
- Flynn's classification (MIMD,SIMD etc.)
- Memory Structures.
- Initial issues in MIMD and SIMD discussion.
- Audio Recording: on MIMD distributed memory architecture.
Lecture (4) foilsets -- September 12, 1996
Lecture (5) foilsets -- September 17, 1996
- Comments: we returned to Laplace equation discussion (which we
had used for performance analysis) starting with foil:
Discretized Form of Laplace'e Equation on a Parallel Processor
in Programming
Models and Performance for Laplace's Equation discussing sequential
and HPF implementation. We briefly mentioned MPI version and
introduced Fortran90 ending with foil:
Important Features of Fortran90
in
HPCC Software Technologies Fall 96 -- Overview and HPF.
- This took us back to the discussion of parallel software in
the context of example of Jacobi Iteration for Laplace equation
in a square box of 256 grid points on 16 processors.
- We already used this example to discuss performance earlier
in the semester.
- HPF was briefly covered as well as brief discussion of MPI
implementation of this example.
- The lecture was finished with initial remarks on Fortran90.
- Audio Recording: no audio for this lecture.
Lecture(6) foilsets -- September 24, 1996
Lecture(7) foilsets -- September 26, 1996
Lecture(8) foilsets -- October 1, 1996
Lecture(9) foilsets -- October 3, 1996
Lecture(10) foilsets -- October 8, 1996
Lecture(11) foilsets -- October 10, 1996
Lecture(12) foilsets -- October 15, 1996
- Comments:
- Audio Recording: < a href="http://www.npac.syr.edu/users/gcf/cps615oct1596/index.html">End of N-Body Discussion and Beginning of Numerical Integration.
Lecture(13) foilsets -- October 17, 1996
- Comments:
- Audio Recording: not available.
Lecture(14) foilsets -- October 22, 1996
Lecture(15) foilsets -- October 24, 1996
- Comments: Two different topics were covered. First, starting with last remarks on Monte Carlo methods from Numerical Integration which includes:
- Monte Carlo Integration for large scale Problems using Experimental and
Theoretical high energy physics as an example.
- This includes accept-reject methods, uniform weighting and parallel algorithms.
- The lecture finally finishes with HPF discussion in foil 23: HPF$ INDEPENDENT, NEW Variable. Embarrassingly parallel DO INDEPENDENT discussed in the Monte Carlo case and HPF2 Changes.
- Audio Recording: End of Numerical Integration and Very End of HPF.
Lecture(16) foilsets -- October 31, 1996
Lecture(17) foilsets -- November 7, 1996
Lecture(18) foilsets -- November 8, 1996
Lecture(19) foilsets -- November 14, 1996
Lecture(20) and (21) foilsets -- November 21 and 25, 1996
- Comments:
- Audio Recording: none available.
Lecture(22) foilsets -- November 26, 1996
Lecture(23) foilsets -- December 5, 1996