基于流式计算的DDoS实时检测方法

TP393.08; 随着大数据时代的到来,DDoS攻击也愈发更具威胁性.近年来,随着以Spark-streaming、Storm为代表的流式计算系统的出现,让DDoS攻击检测不仅可以保持较高的精度,也更加具有实时性.选取三种比较具有代表性的DDoS攻击进行危险建模,构建以Spark-streaming为处理内核、朴素贝叶斯算法为处理算法的实时系统.采用对比实验,验证了本系统具有较高的实时性和准确度....

Full description

Saved in:
Bibliographic Details
Published in计算机应用研究 Vol. 34; no. 3; pp. 892 - 896
Main Author 许承启 何利文 王延松 呼学理 牛小兵
Format Journal Article
LanguageChinese
Published 南京邮电大学计算机学院,南京,210046%中兴通讯股份有限公司,南京,210012 2017
Subjects
DDoS攻击
Spark-streaming
实时性
朴素贝叶斯算法
流计算
DDoS attack
real-time
朴素贝叶斯算法
naive Bayes
实时性
DDoS攻击
流计算
Spark-streaming
Streaming calculation
Online AccessGet full text
ISSN1001-3695
DOI10.3969/j.issn.1001-3695.2017.03.059

Cover

More Information
Summary:TP393.08; 随着大数据时代的到来,DDoS攻击也愈发更具威胁性.近年来,随着以Spark-streaming、Storm为代表的流式计算系统的出现,让DDoS攻击检测不仅可以保持较高的精度,也更加具有实时性.选取三种比较具有代表性的DDoS攻击进行危险建模,构建以Spark-streaming为处理内核、朴素贝叶斯算法为处理算法的实时系统.采用对比实验,验证了本系统具有较高的实时性和准确度.
Bibliography:51-1196/TP
DDoS attack; Streaming calculation; Spark-streaming; naive Bayes; real-time
With the era of big data have arrived,DDoS attack becomes more dangerous. In recent years,streaming calculation system,such as Spark-streaming and Storm etc,have been used in the industry. If this technology can be used to DDoS attack detection. It may be the huge increasing for that. This paper selected three representative ways of DDoS attack as risk modeling,and built a real-time system with Spark-streaming as its processor core and Naive Bayesian as its algorithm. Finally it did the contrast experiments to verify its high real-time performance and high accuracy of this system.
Xu Chengqi1, He Liwen1, Wang Yansong2, Hu Xueli2, Niu Xiaobing2 （1. School of Computer Science ＆ Technology, Nanjing University of Posts ＆ Telecommunications, Nanjing 210046, China; 2. Zhongxing Telecommunication Equipment Corporation, Nanjing 210012, China）
ISSN:1001-3695
DOI:10.3969/j.issn.1001-3695.2017.03.059