CSE8163: Parallel and Distributed Scientific Computing
- Instructor: Ed Luke
- Office: Butler 330 (Engineering Research Center 220)
- Office Hours: 10:00am-11:30am T TH Butler 330 (Or by appointment)
- Meeting Place: Butler 104
- Meeting Time: 2:00pm-3:15pm T TH
- Web Site: http://web.cse.msstate.edu/~luke/Courses/sp18/CSE8163
- Instructor E-Mail: luke@cse.msstate.edu
- Course Topics:
- Properties of Floating Point Computations in Modern Processors
- Properties of Floating Point Arithmetic
- Reproducibility in parallel numerical and scientific computations
- Exact Arithmetic Stategies
- High Performance Techniques for Modern Serial Processors
- Models for Cache Performance and Memory Hierarchy
- Cache Oblivious Algorithms
- Compiler Optimization Issues
- Self Tuning Libraries
- Template Meta-Programming
- Domain Specific Compiler Approaches
- High Performance Techniques for Parallel Processors
- Parallel Models (Programming and Performance)
- Advanced Load Balancing Techniques
- Advanced Adaptive, Dynamic and Irregular Parallel Algorithms:
- Convex Hull
- Graph and Search Algorithms
- Advanced Numerical Methods
- Sparse Linear System Solvers
- Monte Carlo Methods
- Fourier Transform
- Required Text: Introduction to Parallel Computing by Vipin Kumar et. al. (Second edition)
- References
- Grading:
- Literature Review 40% (Paper and Presentation): topics Due February 15th
Topics will be assigned in the second week of class. Individual
work only. The resulting papers will be shared with the rest of the
class.
- Project: 50%
A Term Project will be implemented and a report provided. The results will be presented in a final presentation and paper.
- Project Topic Parallel DSMC, starter program DSMC.tar , example surface meshes: surface_examples.tgz Presentation and report due at time of final exam: 3:00pm-6:00pm Monday, April 30th
- Class Participation 10%
- MSU Honor Code :
- The University Wide Honor Code Policy is a must-reading for all students:
Upon accepting admission to Mississippi State University, a student immediately assumes a commitment to uphold the Honor Code, to accept responsibility for learning, and to follow the philosophy and rules of the Honor Code. Students will be required to state their commitment on examinations, research papers, and other academic work. Ignorance of the rules does not exclude any member of the MSU community from the requirements of the processes of the Honor Code. For more information visit the above link.
- Submission: E-mail attachments sent to luke@cse.msstate.edu
- No Late Submission, submit what you have by the due date.
- Submit all source and include files and makefiles
- Do not submit binaries PLEASE !!!
luke@cse.msstate.edu