摘要
针对当前大规模P2P网络失效检测负载对系统可扩展性的影响，对失效检测结果的共享机制展开了研究，提出了一个基于被动订阅机制的低开销失效检测（L．FD）算法。L-FD算法通过被检测节点来建立检测结果的订阅关系，使每个节点只需保持常数个检测关系即可获得所有邻居节点的失效状态，在节点不发生失效情况下可使检测负载下降为O（N）。该算法的结果共享关系可灵活建立，不受覆盖网拓扑结构及同步时钟等因素的影响，可灵活适应不同的P2P系统。仿真实验及分析结果证实了这一新算法的可行性和有效性。
A study of the failure-detection-result sharing in peer-to-peer （P2P） networks was pertormect to reouce me impact of significant detection overheads on the scalability of large scale P2P systems, and on this basis, a low over- head failure detection （L-FD） algorithm based on the passive subscribing mechanism was proposed. The L-FD al- gorithm can establish the relations of detection results sharing by the monitored nodes. Each node in the system only needs detecting invariable nodes to achieve the status of all neighbors. The L-FD algorithm can reduce the detection overhead complexity to O（N） without failure, in addition it can not be limited by the factors of overlay topology and synchronization when establishing detection-resuh sharing relations, thus, it can be rapidly and flexibly applied to different P2P systems. The experimental results and the corresponding analysis show that the new L-FD algorithm is feasible and effective.
出处
《高技术通讯》
CAS
CSCD
北大核心
2012年第11期1155-1160,共6页
Chinese High Technology Letters
基金
国家自然科学基金（61100029）和高效能服务器和存储技术国家重点实验室开放课题基金（2009HSSA07）资助项目.
关键词
P2P网络
失效检测
结果共享
检测负载
peer-to-peer networks, failure detection, results sharing, detection overhead
