Title :
Progressive construction of consistent global checkpoints
Author :
Garcia, Islene Calciolari ; Buzato, Luiz Eduardo
Author_Institution :
Inst. de Comput., UNICAMP, Saso Paulo, Brazil
Abstract :
A checkpoint pattern is an abstraction of the computation performed by a distributed application. A progressive view of this abstraction is formed by a sequence of consistent global checkpoints that may have occurred in this order during the execution of the application. Considering pairs of checkpoints, we have determined that a checkpoint must be observed before another in a progressive view if the former Z-precedes the latter. Based on the Z-precedence and characteristics of the checkpoint pattern, we propose original algorithms for the progressive construction of consistent global checkpoints. We demonstrate that the Z-precedence between a pair of checkpoints is a much simpler way to express the existence of a zigzag path connecting them, and we discuss other advantages of our relation
Keywords :
distributed processing; program debugging; software fault tolerance; system monitoring; system recovery; Z-precedence; algorithm; checkpoint pattern; computation abstraction; consistent global checkpoints; distributed application; progressive construction; zigzag path; Checkpointing; Debugging; Joining processes; Monitoring; Protocols;
Conference_Titel :
Distributed Computing Systems, 1999. Proceedings. 19th IEEE International Conference on
Conference_Location :
Austin, TX
Print_ISBN :
0-7695-0222-9
DOI :
10.1109/ICDCS.1999.776506