Winter 1996, CMPSC-110B, 4 units, Code 09528
Parallel Scientific Computing

Instructor: Prof. Tao Yang, Department of Computer Science (tyang@cs.ucsb.edu, office Eng I 2151, phone 4384.) Office hours: TR 3:00-3:30pm, 4:50pm-5:20pm.

TA: Maria-Cristina Marinescu, mariacri@cs.ucsb.edu.

Location and time: Lec. TR 3:30-4:45, Phelp 2518. Dis. F1:00-1:50, Engr 1132.

Text Book: Parallel Computing: Theory and Practice, Michael Quinn, McGraw-Hill, 1994.

The class reader is available in the Alternative copy shop under name CS-110B (6556 Pardall Road, Isla Vista, Phone 968-1055).

Reference books (reserved in the library): Richard L. Burden, J. Douglas Faires. Numerical analysis, 4th ed. Boston : PWS-KENT Pub. Co., c1989.

Introduction to Parallel Computing, Vipin Kumar, A. Grama, A. Gupta, G. Karypis, Benjamin/Cummings Pub, 1994.

Tentative Schedule:

• Introduction to scientific computing. Parallel Processing. Design of parallel programs. (Chap 1, Reader Part II., Week 1 of Jan).

  TA (Jan 5): Review the concepts of algorithm, how to count operations, basic complexity analysis. Review Taylor expansion, linear algebra.

  Collect names of students and social security numbers for the Meiko and INTEL accounts.

  (Jan 12) Explain the parallel solutions to the prime-finding problem (Text book Chapter 1). Explain the speedup and execution time figures (1-11, 1-12).

• Parallel programming on the UCSB Meiko CS-2 machine. ( Reader Part I., Week 2 of Jan)

  TA (Jan 19): Discuss how to use the Meiko CS-2 in lab, demonstrate the login, show the parallel machine, compile and run a program.

• Parallel Architecture, communication operations. Performance evaluation. ( Reader Part III., Chapters 3. 4.5, 5.1.1, 5.3.1, Week 3, 4 of Jan.)

  TA (Jan 25): explain an example of parallel programming on the Meiko CS-2.

• Issues in Parallel Programming. Matrix computation. (Reader Part II/III, Week 1, 2 of Feb).

  TA (Feb 2, 9, 16): Review the concept of matrix multiplication, direct and iterative methods for solving a linear system. Complexity.

• Dense and sparse linear systems. ODE/PDEs. (Chapter 9, Reader Part II/III., Week 3, 4 of Feb and Week 1, 2 of March).

  TA (March 23, 1, 8, 15 ): Explain an example of parallel Gaussian Elimination, how to access INTEL Paragon. Review the homework solutions.

• Midterm (Feb 15). Final (March 21, 4-7pm).

Grading: Six homework/programming assignments (40%), Midterm (25%), Final (35%).