| S. Crane. Recovery from Failure in an ObjectOriented Distributed System. Master's thesis, Department of Computer Science, Trinity College Dublin, Ireland, February 1990. |
....COMANDOS [11] project (Esprit project 2071) Before the work described in this thesis had been done, Amadeus had no support for fault tolerance. There had been work done in a predecessor to Amadeus to support faulttolerance using a message logging mechanism to restore consistency after a failure [12]. It was decided not to take this approach in Amadeus because the overheads imposed by this mechanism affect all applications and not just those applications that require reliability. Since transactions are a well understood mechanism and provide a means of allowing programmers to choose which ....
Stephen Crane. Recovery from failure in an object-oriented distributed system. Master 's thesis, Dept. of Computer Science, Trinity College Dublin, May 1990.
....use of resources during normal operation. This section itemizes assumptions about failure in the system and describes the mechanisms for (a) detection of, and recovery from failure and (b) maintaining the data needed to perform the recovery. Further details, and there are many, can be found in [5]. 4.1 The failure model As might be inferred from x1.1, any failure model must necessarily be incomplete as a descriptive system by virtue of Godel s Incompleteness theorem [9] Although its complexity may be arbitrary, disasters are nonetheless ineradicable. 7 A limiting factor on the ....
S. Crane. Recovery from Failure in an ObjectOriented Distributed System. Master's thesis, Department of Computer Science, Trinity College Dublin, Ireland, February 1990.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC