CSE 260 - Parallel Computation (Fall 2002)

Course Description



Parallel computation has a rich and varied history, and the techniques for solving problems on high performance parallel computers are both intriguing and intellectually appealing.

This course will cover principles and practices of parallel programming with applications to real world problems. There will be a fair bit of programming in the course, which will culminate in a mini research project. Our computational platform will be a 32 processor Linux Cluster operated by Academic Computing Services.

The prerequisite for CSE 260 is graduate standing. I recommend that you have a background in graduate architecture or operating system. This background may be met by taking CSE 240A or CSE 221, respectively, or an equivalent course.

The home page for this course is http://www.cse.ucsd.edu/~baden/classes/cse260_fa02/.

The course has one required text: Parallel Programming with MPI, by P. Pacheco, Morgan Kaufmann Publishers, 1997.


Syllabus

  • Fundamentals (motivation, execution models, address space organization, performance)
  • Programming (message passing, threads, higher level models)
  • Data structures and their efficient implementation (hashed oct-trees, multi-resolution representation, graphs, partitioning)
  • Applications (sorting, sparse problems, discretization)

    • See the web page for the  last offering of CSE 260 (Spring 2001).

    Maintained by Scott B. Baden. Last modified: Sunday, August 11, 2002 06:51 AM