论文部分内容阅读
包分类技术是路由器和防火墙等网络设备的关键技术.在高速网络环境中,快速、准确地将数据包进行分类对于网络发展具有重要意义.提出一种基于流的局部特性和多级查找的高效包分类算法,同时可以支持规则库动态更新,实现快速包分类.该算法分为三级结构,第一级缓存用于存放最近10秒内到达的流,第二级计数布鲁姆过滤器存放最近10秒至60秒内到达的流,第三级计数布鲁姆过滤器存放剩余的流.实验表明:该算法比传统的包分类算法,在消耗内存接近的情况下具有更好的时间性能.
Packet classification technology is the key technology of network equipment such as routers and firewalls.In the high-speed network environment, the classification of data packets quickly and accurately is of great significance for network development.This paper proposes a new algorithm based on local characteristics of flow and multi-level lookup Packet classification algorithm.At the same time, it can support dynamic update of rule base and realize rapid packet classification.This algorithm is divided into three levels structure, the first level cache is used to store the flow which arrives in the last 10 seconds, the second level counts Bloom filter store The last arriving stream in 10 seconds to 60 seconds, and the third-level full bloom filter in the remaining stream.Experiments show that this algorithm has better time performance than the traditional packet classification algorithm when the memory consumption is close to .