Date of Award
2014
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Computer Science
Committee Chair
Daniel Rochowiak
Committee Member
Harry S. Delugach
Committee Member
Letha Etzkorn
Subject(s)
Computer software--Development, Object-oriented programming (Computer Science), Error correcting codes (Information theory)
Abstract
As computer languages gained the ability to reflect upon their running states, it should be possible for programs produced using those languages to detect and correct errors. In this thesis, I propose a system that is capable of accomplishing such a task. I created a system that is capable of detecting and correcting exceptional behavior within a running process. The system uses an aspect-oriented programming framework to inject advices into methods. These advices catch exceptional behavior, gather information about the state of the method, and pass the information to an external process for evaluation. Once the external process is able to find a common cause for the errors, it sends back a procedure to prevent the exception from occurring. I tested this system to prove that it provides minimal overhead to the system. Testing also shows that the system can provide a radical performance improvement in highly error prone environments.
Recommended Citation
Gardner, Christopher J., "An architecture for self-healing software systems" (2014). Theses. 67.
https://louis.uah.edu/uah-theses/67