【摘要】：提出用分布式哈希表(DHT)為每臺云服務器產生一個唯一的節(jié)點編號,該編號作為網(wǎng)絡拓撲結構、檢索信息存儲和信息查詢共同的標志符,從而形成一個適合分布式計算的結構化P2P覆蓋網(wǎng)。設計了新的拓撲和路由協(xié)議來解決云資源的常數(shù)跳定位問題。仿真實驗表明,經典的P2P算法平均查找跳數(shù)與網(wǎng)絡規(guī)模成正相關,無法依據(jù)云計算的實際需要人為地控制查找跳數(shù);該算法的平均查找跳數(shù)與網(wǎng)絡規(guī)模無關,隨著網(wǎng)絡規(guī)模的增大而趨向于設定值,可以解決云資源的常數(shù)跳定位問題。
[Abstract]:A distributed hash table (DHT) is proposed to generate a unique node number for each cloud server, which is used as a network topology to retrieve information storage and common identifier for information query. Thus, a structured P2P overlay network suitable for distributed computing is formed. A new topology and routing protocol are designed to solve the problem of constant hop localization of cloud resources. The simulation results show that the average search hops of the classical P2P algorithm are positively correlated with the network size and can not be artificially controlled according to the actual needs of cloud computing. The average search hops of this algorithm are independent of network size and tend to set values with the increase of network size, which can solve the problem of constant hop localization of cloud resources.
【作者單位】： 上海理工大學光電信息與計算機工程學院;上海理工大學網(wǎng)絡與信息中心辦公室;上海理工大學經濟與管理學院;
【基金】：國家自然科學基金資助項目(61170277) 上海市教委科研創(chuàng)新重點資助項目(12zz137)
【分類號】：TP393.02
，

本文編號：2347334