READ：一种三通信级分布式超点检测算法
READ: a three-communicating-stage distributed super points detections algorithm
摘要：
超级点是在一段时间内与网络中大量对等方交互的主机。超点检测在网络研究和应用中占有重要地位。随着网络规模的增大，分布式超点检测成为研究的热点。与单节点超点检测算法相比，多节点分布式环境下超点检测的难点在于如何减少通信开销。为此，本文提出了一种三阶段通信分布式超点检测算法：基于粗估计的异步分布式超点检测算法（READ）。READ使用轻量级的估计器Rough Estimate（RE），其计算速度快，生成候选超点占用的内存少。同时，利用线性估计器（LE）精确估计每个候选超点的基数，从而正确检测超点。在READ中，每个节点异步扫描IP地址对。当到达时间窗口边界时，READ开始三级通信以检测超点。本文证明了分布式环境下READ的准确性不低于单节点环境下的准确性。使用四组10 Gb/s和40 Gb/s的真实高速网络流量来测试READ。实验结果表明，READ算法不仅在分布式环境下具有较高的准确率，而且与现有算法相比，其通信负担小于5%。
英文摘要：
A super point is a host that interacts with a far larger number of counterparts in the network over a period of time. Super point detection plays an important role in network research and application. With the increase of network scale, distributed super point detection has become a hot research topic. Compared with single-node super point detection algorithm, the difficulty of super point detection in multi-node distributed environment is how to reduce communication overhead. Therefore, this paper proposes a three-stage communication distributed super point detection algorithm: Rough Estimator based Asynchronous Distributed super point detection algorithm (READ). READ uses a lightweight estimator, the Rough Estimator (RE), which is fast in computation and takes less memory to generate candidate super point. At the same time, the Linear Estimator (LE) is used to accurately estimate the cardinality of each candidate super point, so as to detect the super point correctly. In READ, each node scans IP address pairs asynchronously. When reaching the time window boundary, READ starts three-stage communication to detect the super point. In this paper, we proof that the accuracy of READ in distributed environment is no less than that in the single node environment. Four groups of 10 Gb/s and 40 Gb/s real-world high-speed network traffic are used to test READ. The experimental results show that READ not only has higher accuracy in distributed environment, but also has less than 5% of communication burden compared with existing algorithms.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
