Fault Tolerant Systems
Teaching > Graduate Courses
Fault Tolerant Computer Systems
Some applications require high dependability computer systems. Such applications include but are not limited to transaction, medical and space program systems. Systems designated for such applications must be designed under different specifications and limitations. The undergraduate program of our Department gives the opportunity to our students to learn how to design systems aiming performance, area, power consumption or some combination of them. This course provides insight on the design rules they must follow when the primary aim is dependability (reliability, availability, maintainability, ...).
The lectures given are based on :
Basic Reading
- Design and Analysis of Fault - Tolerant Digital Systems, Barry W. Johnson, Addison-Wesley Publishing Company, 1989.
Additional Readings
- Theory and Practice of Reliable System Design, D. P. Siewiorek and R. Swarz, Digital Press, 1982.
- Fault Tolerance Principles and Practices, T. Anderson and P.A. Lee, Prentice-Hall International, 1981.
- Fault Tolerant and Fault Testable Hardware Design, P. K. Lala, Prentice-Hall International, 1985.
- Error Coding for Arithmetic Processors, T.R.N. Rao, Academic Press, 1974.
- Error Correcting Codes, W.W. Petersen and E.J Weldon Jr., 2nd edition, MIT Press, 1972.
- Fault-tolerant Systems, Israel Koren & C. Mani Krishna, Morgan Kaufmann, 2007.
An outline of the subjects covered during the lectures is :
Introduction
- Dependability as a Design Goal
- Applications
- Terminology, Complexity and Models
Faults, Errors, Failures
- Physical Causes of Faults
- Fault Characteristics
- Common Fault Models
- Common Error Models
Design Techniques
- Hardware Redundancy
- Information Redundancy
- Time Redundancy
- Software Redundancy
Application of the Developed Theory
- Design Cycle Modification to accommodate Dependability
- A Design Example
Marks are based on a written test.