大规模网络中基于HDHT的病毒疫苗分发算法

摘要

目前,反病毒系统防御病毒的主要方法是依靠反病毒中心服务器定期更新各个节点上的病毒定义文件和安装系统补丁,这种被动应对策略常常不够及时,且容易使服务器节点成为系统的性能瓶颈,具有较大的滞后性,也不能适应大规模网络系统.针对大规模网络环境提出了一种高效的基于层次式分布散列表(hierarchicaldistributed Hash table,HDHT)的反病毒疫苗分发算法.算法基于一种新型层次式分布散列表网络拓扑结构HDHT,利用网络边缘节点直接通信来提高病毒疫苗分发效率,从而有效分担服务器负载.为了增强算法的性能,算法还引入一种简洁的节点信誉评估机制.从分发效率、系统开销等方面,对病毒疫苗分发算法的性能进行实验验证和性能分析.结论表明,该算法可有效实现病毒疫苗在网络中快速分发和部署,同时减轻了网络和服务器的负载.%In order to defend viruses, current anti-virus systems usually deploy central servers to update virus definition files and install system patches on each client node regularly. This kind of passive strategy is often not timely, and the central server is easy to be the performance bottleneck of the system, which is obviously not suitable for large-scale networks. An efficient virus vaccine distribution algorithm based on hierarchical distributed Hash table (HDHT) for large-scale networks is proposed. HDHT is one of our research results, which is a novel HDHT network topology. The algorithm utilizes the direct communication between network edge nodes to speed up the vaccine distribution, which effectively shares server loads. In order to enhance the performance of the algorithm, a simple reputation evaluation mechanism is also introduced. Experimental results and performance analysis from the aspects of the distribution efficiency and system workloads show that the proposed vaccine distribution algorithm can reduce network and server loads with rapid distribution speed of vaccines.