Vulnerabilities / Threats
8/16/2010
01:37 PM
50%
50%

Passwords Quickly Hacked With PC Graphics Cards

Georgia Tech researchers find that high-end, readily available graphics processing units are powerful enough to easily crack secret codes.




Slideshow: Cloud Security Pros And Cons
(click for larger image and for full photo gallery)
Passwords with fewer than 12 characters can be quickly brute-force decoded using a PC graphics processing unit (GPU) that costs just a few hundred dollars, according to researchers at the Georgia Institute of Technology.

"We've been using a commonly available graphics processor to test the integrity of typical passwords of the kind in use here at Georgia Tech and many other places," said Richard Boyd, a senior research scientist at the university's research institute, in a statement. "Right now we can confidently say that a seven-character password is hopelessly inadequate."

Today's top graphics processors offer about two teraflops of parallel processing power. For comparison, "in the year 2000, the world's fastest supercomputer, a cluster of linked machines costing $110 million, operated at slightly more than 7 teraflops," he said.

The barrier to using multi-core graphics processors -- available from Nvidia or AMD's ATI division -- for compute-intensive processes other than graphics processing, said Boyd, first fell in 2007, when Nvidia released a C-based software development kit. "Once Nvidia did that, interest in GPUs really started taking off," he said. "If you can write a C program, you can program a GPU now." Or use it to crack a password.

Furthermore, thanks to Moore's Law, graphics processors continue to increase in power, which means that GPUs will get better, not worse, at cracking passwords.

But who needs a graphics processor? People often create and rely on simple passwords, and many websites use passwords more for psychological than security purposes.

But the Georgia Tech research underscores the importance of getting people to adopt longer, non-simple passwords to make them safer against attack. "Length is a major factor in protecting against 'brute forcing' a password," according to one research scientist involved in the project, Joshua Davis. "A computer keyboard contains 95 characters, and every time you add another character, your protection goes up exponentially, by 95 times."

For the record, to defend against GPU attacks, the password researchers recommend using sentence-length passwords that mix letters with numbers or symbols, and which are at least 12 characters long.

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
ANON1241631011972
50%
50%
ANON1241631011972,
User Rank: Apprentice
11/1/2011 | 6:35:42 PM
re: Passwords Quickly Hacked With PC Graphics Cards
I think this article overstates the capabilities and the use case does not generally exist in the real world. Passwords do not exist in a cyberspace vacuum, just waiting to be attacked in isolation. They exist in combination with a user identifier and a challenge/response system behind a UI. While the brute force methods may be capable of generating all the possible combinations, they still have to test them against a validation challenge with the correct response set. Most password challenge systems have two additional inhibitors against brute force attacks: 1. They don't respond at GPU speeds 2. They usually disable the account after 3 to 5 failed attempts. So, unless the brute force program has the password database available and the appropriate decode algorithms to provide the correct challenge/response patterns, it will not succeed. If the attacker has that kind of access, he or she is already past the firewall and well into the bowels of the network.
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-2014-3971
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-2004-2771
Published: 2014-12-24
The expand function in fio.c in Heirloom mailx 12.5 and earlier and BSD mailx 8.1.2 and earlier allows remote attackers to execute arbitrary commands via shell metacharacters in an email address.

CVE-2014-3569
Published: 2014-12-24
The ssl23_get_client_hello function in s23_srvr.c in OpenSSL 1.0.1j does not properly handle attempts to use unsupported protocols, which allows remote attackers to cause a denial of service (NULL pointer dereference and daemon crash) via an unexpected handshake, as demonstrated by an SSLv3 handshak...

CVE-2014-4322
Published: 2014-12-24
drivers/misc/qseecom.c in the QSEECOM driver for the Linux kernel 3.x, as used in Qualcomm Innovation Center (QuIC) Android contributions for MSM devices and other products, does not validate certain offset, length, and base values within an ioctl call, which allows attackers to gain privileges or c...

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.