The 100Gbps DDoS Attack That No One Saw

Not many people noticed, but late last week one of the largest distributed denial of service (DDoS)  attacks in the history of the Internet was taking place—a massive nine-hour barrage that leveled an unrelenting 100 Gigabit/s (Gbps) of traffic at its peak.

The attack took place on 24 September, and to date the victim of the attack is remaining in the shadows, not wanting to be publicly identified. The target website is protected by cloud security vendor Incapsula, which was able to withstand the massive distributed denial-of-service (DDoS) attack and keep the targete site up and running.

Raw bandwidth, not DNS reflection

Incapsula co-founder Marc Gaffan explained to eWEEK that the attacked site is in an industry that is constantly under assault. The attack leveraged raw bandwidth under the control of the attacker and was not a DNS reflection or amplification attack, Gaffan said. In March of this year, another 100 Gbps attack was reported that leveraged DNS reflection. With DNS reflection, the number of inbound connections to a target Website is amplified by taking advantage of poorly configured DNS servers.

Gaffan noted that in a DNS reflection attack, the actual attack bandwidth can be amplified to deliver by a factor of 30.

“The most outstanding thing about this attack is that it did not use any amplification, which means that they had 100 Gbps of available bandwidth on their own,” Gaffan said. “The attack lasted nine hours, and that type of bandwidth is not cheap or readily available.”

Gaffan added that he was shocked that 100 Gbps of bandwidth was being used in a targeted attack and — other than Incapsula and its own service providers that were on the receiving end — no one seemed to notice.

Attacker had limits

As to why the attack lasted nine hours and not longer, Gaffan suspects it is only because the attacker had hit its own limit.

“In a typical DDoS, an attacker will turn up the gauges on bandwidth as much as possible until they are able to break down a site,” Gaffan said. “There is no point in throwing 100 Gbps of traffic against a site that will go down with 10 Gbps.”

Gaffan’s assumption is that 100 Gbps of raw bandwidth was the attacker’s own physical limit of capacity. The attack did not start off with 100 Gbps of traffic, but rather that was the peak bandwidth achieved during the nine-hour attack, he said.

“Our assumption is the attacker realised that they were getting blocked and that there was no point in continuing to throw rocks at a tank if the rocks were just going to bounce off,” Gaffan said.

How to survive an attack

There are a number of reasons why Incapsula was able to withstand the 100 Gbps attack. For one, Incapsula has more than 400 Gbps of capacity that is globally distributed around the world, according to Gaffan. Incapsula also has its own Web application firewall (WAF) and associated DDoS protection technologies to further limit the risk.

Throughout the nine-hour ordeal, Gaffan said there was no network slowdown for Incapsula customers.

While Incapsula was able to repel this most recent attack, Gaffan cautions that the attack could have been much bigger and there have been some key takeaways from the experience.

“The biggest lesson is that we need to continue to build a bigger boat,” he said.

Attacks are always getting bigger, and an attack 50 times larger than the one his company defended against might be looming just over the horizon, Gaffan warned.

“The abundance of attack capacity that exists is growing in orders of magnitude, month after month,” he said.

Sean Michael Kerner is a senior editor at eWEEK and InternetNews.com. Follow him on Twitter @TechJournalist.

Originally published on eWeek.