Supported by the National Natural Science Foundation of China under Grant Nos.60433040,60603070(国家自然科学基金);the China Postdoctoral Science Foundation under Grant No.2005038064(中国博士后基金)
A Hybrid Peer-to-Peer Lookup Service Algorithm on Distributed Hash Table
Author:
Affiliation:
Fund Project:
摘要
|
图/表
|
访问统计
|
参考文献
|
相似文献
|
引证文献
|
资源附件
|
文章评论
摘要:
使用分布式哈希表(distributed hash table,简称DHT)的应用系统必须在O(1)发现算法和O(logN)发现算法系列中选择适应的DHT协议.但是,不同网络波动程度的应用场景要求理想的DHT协议根据网络波动率能够自适应地调整.提出一种发现算法ROAD(routing on active and demand),在延时和波动率之间自适应地调整以提供更好的性能.设计ROAD的关键挑战是构建路由表和降低延时的算法.通过构建加速路由表,加快发现服务的速度,降低消息转发的延时,并通过幂次序组播算法
Abstract:
An application using a distributed hash table (DHT) with N nodes must choose a DHT protocol from the spectrum between O(1) lookup protocols and O(logN). However, various applications under different network churns require that an idea DHT would be adaptive in according with the churn rates. ROAD (routing on active and demand), a new lookup algorithm, adjusts itself to provide the best performance across a range of lookup delay and churn rates. The key challenges in the design of ROAD are the algorithms that construct the routing table’s size and decrease the delay. It will speed up the lookup process and reduce the service delay with the expressed routing table and power sorting multicast algorithm. Simulations show that ROAD maintains an efficient lookup delay versus churn rate tradeoff than the existing DHTs. ROAD should be expanded into a mechanism that provides some kinds of lookup services with a range of qualities of service through super-peers choosing methods.