| 摘要 |
PROBLEM TO BE SOLVED: To continue event-driven P2P real-time communication even when a master ends abnormally by using another peer as a master. SOLUTION: Architecture is configured such that, according to past reliable events distributed from a master peer 4 of a shared space, a new master peer 1 of the shared space is selected uniquely. All peers 2 and 3 can uniquely determine an alternative master peer candidate according to a master candidate list (master candidate list generation and holding part 16) created with redistributed events. A master alternation notification from the peers 2 and 3 other than the master peer 4 can be authenticated. Even upon an abnormal end of the master peer 4 during event redistribution, a final event is redistributed after the master alternation to keep the same state of the peers 2 and 3. COPYRIGHT: (C)2006,JPO&NCIPI
|
