DDoS is the acronym for Distributed Denial of Service.

The telephone system, computer system and Domain Name System (DNS) sometimes become unusable during peak hours because of supply and demand. However, when an intruder or hacker interrupts the system, takes control of the computer, prevents the legitimate user from using it, and forces the computer to send such a large amount of email to another person that it cannot be handled by the recipient's save disk, a Denial of Service (DoS) attack happens. If an intruder attacks a particular computer, takes control of it, sends extraordinary amount of data to a website and distributes it to numerous email addresses affecting the computer network, the intrusion is called a Distributed Denial of Service attack.[1]

Background

The CERT/CC at Canegie Mellon University documented the first incident of Denial Of Service Attack in 1999 when the Trinoo and Tribe Flood Network (TFN) DDoS Network tools were widely distributed. The two DDoS used UDP Flood attack, TCP SYN flood, ICMP echo request flood, and ICMP directed broadcast denial of service attacks respectively.[2] Trinoo attacked a single computer from Minnesota University, affected around 227 systems, and became unusable for more than two days.[3]

On February 2000, a massive DDoS attack paralyzed high profile websites including Yahoo!, Buy.com, eBay, CNN, Amazon.com, ZDNet.com, E-Trade, and Excite, which together lost an estimated amount of $1.7 billion. A suspect, a Canadian juvenile with the online alias "mafiaboy," was arrested on April of the same year. He pleaded guilty on January 18, 2001 on 56 charges of mischief and illegal use of computer services.[4]

Over the years, intruders have used different DDoS tools to affect computer systems:

  • Stacheldraht, 1.666 DDoS tool was discovered and widely spread on multiple compromised hosts in several organizations;[5]
  • Love Letter Worm, a malicious VBScript which was spread through emails, Windows file sharing, IRC, USENET news and through webpages affecting more than 500,000 computer systems;[6]
  • T0rnkit, also distributed by intruders using six different versions of rootkit;[7]
  • W/32/Sircam, an e-mail-borne virus;[8]
  • Leaves, which was capable of updating and changing its functionality during a hack, affected millions of internet users in five Chinese provinces when an unknown hacker attacked the the server of DNSPod, a Chinese domain name registrar in 2009;[9]

as well as many other viruses and worms distributed by hackers to cripple computer networks in homes and organizations.

Network Solutions spokesperson Shashi Bellamkonda reported that the company experienced a consecutive DDoS attacks on June 20-21, 2011 wherein its costumers were unable to access the server and e-mail and the website became unstable. The company resolved the problem as quickly as possible.[10]

Packet Flooding Attack

The Packet Flooding Attack is the most common type of Denial of Service Attack. The modus operandi of intruders is sending more than acceptable number of packets to a particular destination which consumes the entire bandwidth resources. There are several types of packets used by Packet Flooding Attack tools, including:

  • TCP Floods - SYN, ACK and RST flags are sent to the victim's IP Address
  • ICMP echo request reply (Ping Floods) - A stream of ICMP is sent to the victim's IP Address
  • UDP Floods - A stream of UDP is sent to the victim's IP Address

These attack tools change the characteristics of packets in the packet stream. For example, the Source IP Address is changed to hide the real source of the packet stream. The method of sending packet streams to one or more intermediate sites to create responses that will be sent to a victim is called IP Spoofing.[11] Other packet stream attributes that are altered by intruders are the Source/Destination Ports and Other IP Header Values.

Frequent Targets of Intruder Attacks

According to the CERT report, "Trends in Denial Service Attack Technology," the most frequent targets are Windows end-users and Internet Routing Technology. An intruder's primary intention in conducting DoS attack is to prevent the use of computer or network resources. A computer controlled by a hacker is known as "zombie" or "bot," while a controlled computer network is referred as a "botnet" or "zombie army."[12]

Reasons Why Internet is Vulnerable to Attacks

Internet-connected systems are still vulnerable to DoS attacks despite active security efforts because of the following reasons:

  • The Internet is composed of limited and consumable resources
  • Internet security is highly interdependent[13]

Developments/Researches on DDoS Attacks

Neustar SiteProtect

Verisign Research on DDoS Attacks

On May 2011, a new research commissioned by Verisign found a widespread DDoS attacks on businesses in all industries and they lack adequate protection against it. The research found out that 63% out of the 225 IT decision-makers who respond to the survey reported that they experience more than one DDoS attacks for the past year, 11% said they experienced more than 6 attacks. Sixty seven percent (67%) of the respondents believed that the frequency of DDoS attacks within the next two years will increase or stay the same and 71% of the respondents believe that DDoS protection is important to maintain their website and services available and 71% of the respondents who lack DDoS protection plan to implent solutions within the next 12 months. [14]

References