New Page 1

CprE 545 Fault Tolerant Systems

Important Dates

Proposal Due: February 28 or earlier

Final Report Due: April 30 or earlier

Scope: It is your responsibility to find a relevant topic for your term paper/project. If you are involved in research or development in this area, your project can relate to your work as long as your work can be public. If you prefer to perform some original work, that is very welcome. The project is to study the selected problem and quality and shortcomings of available solutions. In the report, you must also include any new solution you can think of and general comments on why the problem is important. It is important to note that it takes about two weeks to write a good report. If you need any assistance in selecting a topic, feel free to contact me. I have an open office hour policy. I can only help and you have to work to succeed.

Proposal: The project proposal should be about one page long and must be submitted by email (text or ps file). It must include the title of the area you will research, possible title of the report, selected papers or other literature, and nature of research you will carry out.

Final Report: The final report should be written following the format of a research paper. The length should be about 25 pages double spaced. (The length is only a guideline and you take what it takes to write a good paper. Please use 12 point font and keep it double spaced.)

Grading: The projects will be graded based on the following factors: Technical Content, Originality, Significance, Contribution, and Quality of Presentation.

Suggested Topics: Here is a list of Possible Term Paper Topics:

	Faults, errors, and failure models
	Redundancy management techniques
	Dependability evaluation techniques
	Evaluation criteria and reliability estimation
	Testing and Built-in-Self-Test Functional testing
	System-level testing
	Self-checking and fail-safe circuits
	System-level fault diagnosis
	Distributed diagnosis algorithms
	Designs for maintainability and maintainability techniques
	Issues in design of fault-tolerant systems
	Information redundancy techniques
	Fault resilient and fault tolerant structures (hypercubes, star graphs, binary trees, meshes)
	Fault-tolerant communication
	Fault Tolerance in multi-stage interconnection network
	Software fault-tolerance
	Issues in reliable software design
	Fault tolerant data structures
	Hardware support for software fault tolerance
	Checkpointing and recovery techniques
	Algorithm-based fault tolerance
	Software safety
	Fault tolerant multiprocessor systems
	Fault tolerance in commercial systems
	Byzantine fault tolerance
	Fault tolerant subsystem design (Processor, Memory, I/O, Clocks)
	Synchronization and initialization in fault tolerant systems
	Any other topic I missed and you find interesting and related