EE 686 - 001
Advanced Computer Architecture Design
Course Syllabus
Spring 1998

Instructor : Dr. J. Robert Heath

Office : 310 EEX (257- 3124), E -Mail : heath@engr.uky.edu

Office Hours : M ( 10:00 - 11:00 AM )
W ( 3:00 - 4:00 PM)
F (11:00 - 12:00 noon)

Meeting Schedule : MWF (2:00 - 2:50PM) AH 253

Texts : 1. J. Hennessy and D. Patterson, Computer Architecture: A Quantitative Approach, Second Edition, Morgan Kaufmann, 1996.

2. D. Culler and J.P. Singh, Parallel Computer Architecture: A Hardware/Software Approach, Morgan Kaufmann, 1998.

References : 1. Richard Y. Kain, Advanced Computer Architercture: A Systems Design Appraoch, Prentice Hall, 1996.

2. K. Hwang and Z. Xu, Scalable Parallel and Cluster Computing: Architecture and Programming, McGraw Hill, 1998.

3. T.L. Casavrant, P. Tvrdik, and F Plasil, Parallel Computers: Theory and Practice, IEEE Computer Society Press, 1996

4. Dan I. Moldovan, Parallel Processing: Fom Applications To Systems, Morgan Kaufmann, 1993.

5. Kai Hwang, Advanced Computer Architecture: Parallelism, Scalability, Programmability, McGraw Hill, 1993.

Course Description : A study of current diverse advanced architectures and issues such as microprogrammed, instruction/process level parallel architectures, array and vector, networked, dataflow, and distributed architectures; parallel algorithms and programming; applications and example systems employing parallel architetures; matching applications to architectures; performance evaluation; consideration of architectures and issues of the future.
Topical Outline : 1. Advanced Issues of Memory-Hierarchy Design
2. I/O and Storage Systems
3. Parallel Processing Overview and Models of Parallel Computation.
4. Example Architectures for Parallel Processing (Overview)
5. Parallel Programs
6. Programming for Performance
7. Workload-driven Evaluation
8. Shared Memory Multiprocessors
9. Snoop-based Multiprocessor Design
10. Scalable Multiprocessor
11. Scalable Shared Address Space Multiprocessor
12. Hardware-Software Tradeoffs
13. Interconnection Network Design
14. Latency Tolerance
15. Research Issues and Future Directions
Grade : Two Tests ( 25% each )
Clasroom Presentation (s)
Homework
Final Exam

Your final grade will generally be determined by the number of points you have accumulated from 100 possible points as follows:

A:
B
:
C
:
E
:
90 - 100 pts
80 - 89 pts
70 - 79 pts
Below 70 pts

Final grades may be scaled.

50%
20%
5%
25%
Class Attendance : You are expected to attend all classes. You are responsible for all business conducted in all classes.
Make-up Tests : Make-up tests will only be given for excused absences according to applicable University policies.
Homework : Homework will be periodically assigned and reviewed by the instructor. It will not be graded on a detailed basis.