Department of Computer Science News
Prof. Mikhail Nesterenko Wins Best Paper AwardPosted Aug. 9, 2011
Prof. Mikhail Nesterenko (Kent State University) and Sebastien Tixeuil (University of Paris) were awarded "Best Paper" at the 25th IEEE International Conference on Advanced Information Networking and Applications (AINA-2011) in Singapore.
Their paper, entitled "Ideal Stabilization", introduces a new approach to reasoning about optimistic fault recovery. The concept of ideal stabilization is related to self-stabilization. A self-stabilizing program eventually reaches legitimate (correct) set of states regardless of the state to which a fault pushes the program. Thus, the program eventually recovers, irrespective of the nature of the fault once its influence stops. Classic self-stabilization prescribes only that eventually the program behaves correctly. This potentially limits the applicability of the property as the designers do not have an explicit way of specifying how the program behaves during recovery.
Ideal stabilization states that every possible program state is legitimate. This way, the designers have complete control in specifying how the program should recover from faults. This approach gives formal rigor and practical applicability to the concept of stabilization. Nesterenko and Tixeuil provide several illustrative examples of fault-tolerant programs and prove them ideally stabilizing.