Attacks/Breaches
11/4/2011
01:53 PM
50%
50%

Will Cloud Signaling Contain DDoS Attacks?

Arbor Networks' "Bat-signal" for distributed denial of service attacks culls your network service provider's resources to help stop it.

Is a distributed denial of service (DDoS) attack crippling your corporate network? Then signal your upstream service provider so it can filter out the attack using carrier-grade equipment.

That's the premise of "cloud signaling," which is a set of technical capabilities and mitigation strategies offered by network security vendor Arbor Networks. Announced in May, Arbor released the first related product in August, named Pravail, which it bills as an availability protection system.

A business that's already purchased DDoS defenses can typically mitigate small attacks, but those that flood the network pipe, or target the application level--including intrusion prevention systems and firewalls--may be much more difficult to stop. That's where cloud signaling comes in. "You can either set it to automatically signal your upstream provider, kind of like a Bat-signal--help me--and it will send help, sending all of the [attack] parameters" to the carrier, said Rob Malan, CTO of Arbor Networks, in an interview. Alternately, Pravail users can trigger the signaling manually.

[Businesses aren't ready to defend against the rising tide of sophisticated cyber attacks. Learn more: Advanced Threats Touch Two-Thirds Of Enterprises.]

Calling in the service provider brings carrier-grade packet-scrubbing power to bear on attacks, which means the attacks can be more easily eradicated by the service provider. "So maybe you've got one to 10 gigs of mitigation protection at the enterprise edge, but you have hundreds of gigs of mitigation protection at the carrier," said Malan.

He said the idea for cloud signaling arose during a late-night Denny's discussion he had with his Arbor co-founder, Farnam Jahanian, who now heads the National Science Foundation's computer & information science & engineering directorate. They hit upon the idea of "being able to hit a 'big red button' at the target of the DDoS attack, and signal all the way back upstream to the sources of the attack, automatically protecting the target."

But that discussion happened in 1999; what took 11 years for the idea to become reality? "While we always wanted to do this, you can't make a living selling a promise that it will be great in 10 years. So we needed to get a large install base," said Malan.

"The first major problem was getting the tool into all the carriers' hands so they could do coordinated troubleshooting. The second was getting it into the hands of the enterprise, so they could do coordinated troubleshooting back to the carrier. Once you have that infrastructure covered, you can hook them all together, all the way back to the carrier," he said.

One related step along this path was the Fingerprint Sharing Alliance, launched in 2002 as a way for network providers to share information with each other about large-scale attacks in progress. From there, it wasn't a big leap to having downstream customers share information about attacks that targeted their specific infrastructure.

Recent changes in DDoS attacks have also created an impetus for such information sharing. "The attacks changed in the last couple of years," said Malan. "Some kinds of attacks, you have to see the packets in line, and the only scalable way to do that is at the enterprise or data center edge--you can't sit up in the cloud and look at every packet, it's too packet-intensive; you've got to distribute that to the enterprise."

But when those defenses located at the edge of a business' infrastructure detect an attack, they can't always stop it. Notably, research from Arbor has found that the average DDoS attack bandwidth increased by 102% during 2010, and rose by 1,000% from 2005 to 2010.

When attackers do attempt to DDoS a business, it typically wouldn't take a data center operator long to pick up the phone and call their carrier, seeking a way to mitigate the attack. But with Pravail, said Malan, the process can be entirely automated, and attack signatures passed upstream at the first hint of trouble. That increases the chance that not only will the attack get quickly blocked, but also that the business won't even see an outage.

Comment  | 
Print  | 
More Insights
Register for Dark Reading Newsletters
White Papers
Cartoon
Current Issue
Dark Reading Tech Digest, Dec. 19, 2014
Software-defined networking can be a net plus for security. The key: Work with the network team to implement gradually, test as you go, and take the opportunity to overhaul your security strategy.
Flash Poll
Video
Slideshows
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2014-1449
Published: 2014-12-25
The Maxthon Cloud Browser application before 4.1.6.2000 for Android allows remote attackers to spoof the address bar via crafted JavaScript code that uses the history API.

CVE-2014-2217
Published: 2014-12-25
Absolute path traversal vulnerability in the RadAsyncUpload control in the RadControls in Telerik UI for ASP.NET AJAX before Q3 2012 SP2 allows remote attackers to write to arbitrary files, and consequently execute arbitrary code, via a full pathname in the UploadID metadata value.

CVE-2014-3971
Published: 2014-12-25
The CmdAuthenticate::_authenticateX509 function in db/commands/authentication_commands.cpp in mongod in MongoDB 2.6.x before 2.6.2 allows remote attackers to cause a denial of service (daemon crash) by attempting authentication with an invalid X.509 client certificate.

CVE-2014-7193
Published: 2014-12-25
The Crumb plugin before 3.0.0 for Node.js does not properly restrict token access in situations where a hapi route handler has CORS enabled, which allows remote attackers to obtain sensitive information, and potentially obtain the ability to spoof requests to non-CORS routes, via a crafted web site ...

CVE-2014-7300
Published: 2014-12-25
GNOME Shell 3.14.x before 3.14.1, when the Screen Lock feature is used, does not limit the aggregate memory consumption of all active PrtSc requests, which allows physically proximate attackers to execute arbitrary commands on an unattended workstation by making many PrtSc requests and leveraging a ...

Best of the Web
Dark Reading Radio
Archived Dark Reading Radio
Join us Wednesday, Dec. 17 at 1 p.m. Eastern Time to hear what employers are really looking for in a chief information security officer -- it may not be what you think.