M. Roesler, W. A. Burkhard, and K. B. Cooper, "Efficient Deadlock Resolution for Lock-Based Concurrency Control Schemes," In The 8th International Conference on Distributed Computing Systems, pp. 224--233, IEEE-CS, San Jose, California, June 1988.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....the maximum number of edges from a task or a resource in a GRG is 1. A cycle in a GRG is a necessary and sufficient condition for deadlock. A blocked task T is said to be deadlocked if T is in a cycle or T can only reach deadlocked tasks. Examples of this model can be found in database systems[4, 52]. In the Ada environment, if the resources are non shareable and only one outstanding request is allowed, and no more than two tasks may be involved in either a rendezvous or task termination, the system can be formalized as a Single Resource model. The Mitchell Merritt algorithm[48] is a very ....

.... A Distributed Scheme for Detecting Communication Deadlocks, IEEE Transactions on Software Engineering, Vol. SE 12, No. 4, pp. 531 537, April 1986. 51] R. Obermarck, Distributed Deadlock Detection Algorithm, ACM Transactions on Database Systems, Vol. 7, No. 2, pp. 187 208, June 1982. [52] C. Papadimitriou, The Theory of Database Concurrency Control, Principles of Computer Science Series, Computer Science Press, Rockville, Maryland, 1986. 53] M. Roesler and W. A. Burkhard, Resolution of Deadlocks in Object Oriented Distributed Systems, IEEE Transactions on Software ....

M. Roesler, W. A. Burkhard, and K. B. Cooper, "Efficient Deadlock Resolution for Lock-Based Concurrency Control Schemes," In The 8th International Conference on Distributed Computing Systems, pp. 224--233, IEEE-CS, San Jose, California, June 1988.

....to a cycle in the WFG. Therefore, algorithms for this model declare a deadlock when a cycle of waiting transactions is determined. The cycle is resolved if one of the involved transactions is aborted, thereby releasing its locks. Numerous algorithms have been proposed for this model [Obe82,RBC88,Bad86,SH89,MM79,KS91] Some of these algorithms will be described in Section 4. In a computational model in which a transaction can send more than one request at a time and has to wait until all of them are granted, deadlocks are described by the AND (or resource) model. This model applies, for ....

....easily be extended for deadlock detection in a model in which a transaction can issue multiple requests at a time. Since the models are so similar sometimes the authors claim that their algorithms are for the AND model, although the computational model they describe is a single resource one, e.g. RBC88] while others make restrictions to their model basically reducing it to a single resource model [Obe82] An algorithm for deadlock detection in the AND model is given in [LK95] but it is not clear how deadlocks can be resolved with this algorithm. Algorithms especially designed for DBMSs ....

[Article contains additional citation context not shown here]

M. Roesler, W. A. Burkhard, and K. B. Cooper. Efficient deadlock resolution for lock-based concurrency control schemes. In Proc. 9th Int. Conf. on Distributed Computing Systems, pages 224--233, 1988.

....corresponds to a cycle in the WFG. Therefore, algorithms for this model declare a deadlock when a cycle of waiting transactions is determined. The cycle is resolved if one of the involved transactions is aborted, thereby releasing its locks. Numerous algorithms have been proposed for this model [Obe82, RBC88, Bad86, SH89, MM79, KS91]. Some of these algorithms will be described in Sect. 4. In a computational model in which a transaction can send more than one request at a time and has to wait until all of them are granted, deadlocks are described by the AND (or resource) model. This model applies, for instance, in systems ....

....a transaction can issue multiple requests at a time. Since the models are so 82 N. Krivokapi c et al. Distributed deadlock detection similar, sometimes the authors claim that their algorithms are for the AND model, although the computational model they describe is a single resource one, e.g. [RBC88], while others make restrictions to their model, basically reducing it to a single resource model [Obe82] An algorithm for deadlock detection in the AND model is given in [LK95] but it is not clear how deadlocks can be resolved with this algorithm. Algorithms especially designed for DBMSs ....

[Article contains additional citation context not shown here]

Roesler M, Burkhard WA, Cooper KB (1988) Efficient deadlock resolution for lock-based concurrency control schemes. In: Proceedings of the 8th International Conference on Distributed Computing Systems, San Jose, California, June 13-- 17, 1988. IEEE-CS Press, 1988, ISBN 0-8186-0865-X, pp 224--233

....passing. Deadlock detection resolution is an important problem in a distributed system and much attention has been devoted to it in the past few years. Many distributed deadlock detection resolution algorithms have been proposed, however, most of them either have not given a correctness proof [1, 2, 3] or have given an informal proof by using intuitive operational arguments [4, 5, 6] Intuitive operational arguments are prone to errors, and many of the published algorithms have been found to be incorrect [4, 5, 1, 3] Only rigorous proofs, using as few operational arguments as possible, suffice ....

M. Roesler, W. A. Burkhard, and K. B. Cooper, "Efficient deadlock resolution for lockbased concurrency control schemes," in Proc. 8th Int. Conf. Distributed Comput. Syst., pp. 224--233, 1988.

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