IN-Network DDOS detection

本文探讨了DDoS攻击及其对网络的影响,对比了现有的流量镜像和清洗方法的不足,提出了一种新的6.4Tbps速率的检测方案。通过Hyperloglog sketch算法在数据平面计算目标源数量,当超过阈值时触发防御机制,将流量转发至DDoS防护服务商。P4技术在DDoS检测中表现出高效优势。
摘要由CSDN通过智能技术生成

DDos大家都知道,就是挟持大量的流量密集的访问目标网站让它瘫痪。该如何破解DDos攻击呢?这就是今天要聊的内容。

现有市面上主流方法:

对流量做镜像,分析镜像流量是否为可疑流量,出现问题时进行流量清洗来防御DDoS攻击。

缺点:分析、清洗的时间过长,响应反应缓慢。

新的方法

用6.4Tbps的速率检测可疑与正常流量


如何察觉、抵御DDoS攻击?


察觉DDoS攻击的原理

计算数据平面每个目标的源数量,使用Hyperloglog sketch算法(复杂度为log(logn))来计算数量,用计算出的数值与阈值相比较,当超过阈值后判定产生了DDoS攻击,转发到专门的DDoS厂商来解决问题。

Hyperloglog sketch算法介绍

具体计算方式如图,不再赘述。(HASH算法真是万能)

数据平面上的计数流程:


取得HASH算法的值与自己的预设表进行比较,得出IP源的数据范围。

使用P4方式进行DDoS detection的统计结果


总之,使用P4的方式进行DDoS的detection是十分优越的。

  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
DDoS Attacks: Evolution, Detection, Prevention, Reaction, and Tolerance discusses the evolution of distributed denial-of-service (DDoS) attacks, how to detect a DDoS attack when one is mounted, how to prevent such attacks from taking place, and how to react when a DDoS attack is in progress, with the goal of tolerating the attack. It introduces types and characteristics of DDoS attacks, reasons why such attacks are often successful, what aspects of the network infrastructure are usual targets, and methods used to launch attacks. The book elaborates upon the emerging botnet technology, current trends in the evolution and use of botnet technology, its role in facilitating the launching of DDoS attacks, and challenges in countering the role of botnets in the proliferation of DDoS attacks. It introduces statistical and machine learning methods applied in the detection and prevention of DDoS attacks in order to provide a clear understanding of the state of the art. It presents DDoS reaction and tolerance mechanisms with a view to studying their effectiveness in protecting network resources without compromising the quality of services. To practically understand how attackers plan and mount DDoS attacks, the authors discuss the development of a testbed that can be used to perform experiments such as attack launching, monitoring of network traffic, and detection of attacks, as well as for testing strategies for prevention, reaction, and mitigation. Finally, the authors address current issues and challenges that need to be overcome to provide even better defense against DDoS attacks. Table of Contents Chapter 1 - Introduction Chapter 2 - DDoS, Machine Learning, Measures Chapter 3 - Botnets: Trends and Challenges Chapter 4 - DDoS Detection Chapter 5 - DDoS Prevention Chapter 6 - DDoS Reaction and Tolerance Chapter 7 - Tools and Systems Chapter 8 - Conclusion and Research Challenges
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值