Distributed denial-of-service (DDoS) attacks have been going on for years. But in recent months they seem to have gained much more attention, in part because of high-profile incidents that affected millions of users.
For instance, in late October 2016 a massive DDoS assault on Domain Name System (DNS) service provider Dyn temporarily shut down some of the biggest sites on the Internet. The incident affected users in much of the East Coast of the United States as well as data centers in Texas, Washington, and California. Dyn said in statements that tens of millions of IP addresses hit its infrastructure during the attack.
Just how much attention DDoS is getting these days is indicated by a recent blog post by the Software Engineering Institute (SEI) at Carnegie Mellon University. The post, entitled, "Distributed Denial of Service Attacks: Four Best Practices for Prevention and Response," became SEI's most visited of the year after just two days, said a spokesman for the institute.
To help defend against such attacks, organizations need to understand that this is not just an IT concern.
"While DDoS attack prevention is partly a technical issue, it is also largely a business issue," said Rachel Kartch, analysis team lead at the CERT Division of SEI, a federally funded research and development center sponsored by the U.S. Department of Defense and operated by CMU, and author of the DDoS post.
Fortunately there are steps organizations can take to better protect themselves against DDoS attacks, and Kartch describes these in the post. In general, organizations should begin planning for attacks in advance, because it’s much more difficult to respond after an attack is already under way. “While DDoS attacks can't be prevented, steps can be taken to make it harder for an attacker to render a network unresponsive,” Kartch noted.
To fortify IT resources against a DDoS attack, it’s vital to make the architecture as resilient as possible. Fortifying network architecture is an important step not just in DDoS network defense, Kartch said, but in ensuring business continuity and protecting the organization from any kind of outage.
To help disperse organizational assets and avoid presenting a single rich target to an attacker. organizations should locate servers in different data centers; ensure that data centers are located on different networks; ensure that data centers have diverse paths, and ensure that the data centers, or the networks that the data centers are connected to, have no notable bottlenecks or single points of failure.
For those organizations that depend on servers and Internet presence, it’s important to make sure resources are geographically dispersed and not located in a single data center, Kartch said. “If resources are already geographically dispersed, it is important to view each data center as having more than one pipe to [the] Internet, and ensure that not all data centers are connected to the same Internet provider,” she said.
While these are best practices for general business continuity and disaster recovery, they will also help ensure organizational resiliency in response to a DDoS attack.
The post also describes other practices for defending against DDoS. One is to deploy appropriate hardware that can handle known attack types and use the options in the hardware that can protect network resources. While bolstering resources will not prevent a DDoS attack from happening, Kartch said, doing so will lessen the impact of an attack.
Certain types of DDoS attacks have existed for a long time, and a lot of network and security hardware is capable of mitigating them. For example, many commercially available network firewalls, web application firewalls, and load balancers can defend against protocol attacks and application-layer attacks, Kartch said. Specialty DDoS mitigation appliances also can protect against these attacks.
Another good practice is to scale up network bandwidth. “For volumetric attacks, the solution some organizations have adopted is simply to scale bandwidth up to be able to absorb a large volume of traffic if necessary,” Kartch said. “That said, volumetric attacks are something of an arms race, and many organizations won't be able or willing to pay for the network bandwidth needed to handle some of the very large attacks we have recently seen. This is primarily an option for very large organizations and service providers.”
It’s likely that DDoS attacks will continue to be a major issue for organizations. A 2016 study by content delivery network provider Akamai said these types of incidents are rising in number as well as in severity and duration. The company reported a 125% increase in DDoS attacks year over year and a 35% rise in the average attack duration.
Cyber security executives need to make it a top priority to protect their organizations against DDoS.