Dark Reading is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Attacks/Breaches

3/24/2017
11:30 AM
Connect Directly
Twitter
LinkedIn
RSS
E-Mail
100%
0%

Sandia Testing New Intrusion Detection Tool That Mimics Human Brain

Neuromorphic Data Microscope can spot malicious patterns in network traffic 100 times faster than current tool, lab claims.

A project that started off as a medical study into cerebral palsy in children has yielded a technology that its creators say could help organizations detect cyberthreats 100 times faster than current products.

The technology was developed by Boston-area startup Lewis Rhodes Labs (LRL) and fine-tuned with the active participation of researchers from Sandia National Laboratories and is called the Neuromorphic Data Microscope.

The technology—currently implemented in the form of a PCIe based processing card—can be used to inspect large volumes of streaming data to find patterns that match known bad behavior faster and more cost-effectively than presently possible, according to the two organizations.

Typical intrusion detection systems sequentially compare relatively small chunks of network data against a library of known malicious patterns to spot threats. The Neuromorphic Data Microscopic does the same pattern matching in a much faster and more parallel manner that mimics the way the human brain processes streaming data.

“One way of thinking about it is when you try matching patterns on a computer, it is a more serial process,” says David Follett, CEO and co-founder of LRL. “The brain is massively parallel.”

It streams data – such as the things within an individual’s range of vision – past stored memory in a very efficient way to help the individual identify people, places, or things that are familiar. 

The Neuromorphic Data Microscope takes the same approach to inspecting massive volumes of streaming network data and finding patterns that suggest malicious behavior. It accomplishes in a single processor card the same level of parallelism that would take multiple racks of traditional cybersecurity systems working in parallel to deliver.

In its current form, the technology accelerates complex pattern matching by a factor of over 100 while using 1,000 times less power than conventional cybersecurity systems, Follett says. LRL will soon implement an ASIC version of the data microscope that will be capable of delivering a 10,000 times performance gain over current intrusion detection tools, he says.

Such capabilities will enable far more complex pattern matching and will allow organizations to spot attacks that are easy to miss currently, says Sandia computer systems expert John Naegle.

“We want to run much more complicated and sophisticated rules against our data to detect malicious types of patterns,” says Naegle. Because of the enormous computing resources it would take to run some of these rules, however, Sandia has had to make conscious decisions about what it can and cannot do with its available computing resources.

“This gives us the opportunity to drastically change the way we do cybersecurity,” Naegle says. “Right now tools are expensive, cumbersome, and very CPU-constrained” to allow for the kind of complex pattern matching Sandia has wanted to do. “This technology gives us an entirely different way to look at the problem and an entirely different way to look for suspicious traffic.”

Naegle describes the data microscope as similar in concept to the Snort open-source intrusion detection tool used by many organizations, including Sandia.

Organizations are under increasing pressure to find better and quicker ways of detecting malicious behavior on their networks. Cybercriminals often are able to easily circumvent many pattern-matching, signature-based intrusion detection systems by making relatively small changes to their malware. So capabilities like those claimed by Sandia and LRL could make a bigger difference.

The idea for the data microscope evolved from a mathematical model that LRL researchers developed for comparing the brains of children suffering from cerebral palsy with brains that do not have the disorder. In using the model, the researchers realized they had developed a way of doing computing that mimicked the manner in which a human brain processes information, a description of the technology on LRL’s website noted.

Sandia is using the Neuromorphic Data Microscope for cybersecurity purposes. But it can be used in a wide range of other applications involving the use of massive volumes of streaming data, Follett says. Examples include applications such as image and video processing, consumer data analysis, fraud identification, and financial trading.

Related stories:

 

 

Jai Vijayan is a seasoned technology reporter with over 20 years of experience in IT trade journalism. He was most recently a Senior Editor at Computerworld, where he covered information security and data privacy issues for the publication. Over the course of his 20-year ... View Full Bio
 

Recommended Reading:

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
COVID-19: Latest Security News & Commentary
Dark Reading Staff 8/10/2020
Researcher Finds New Office Macro Attacks for MacOS
Curtis Franklin Jr., Senior Editor at Dark Reading,  8/7/2020
Lock-Pickers Face an Uncertain Future Online
Seth Rosenblatt, Contributing Writer,  8/10/2020
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
Write a Caption, Win an Amazon Gift Card! Click Here
Latest Comment: They said you could use Zoom anywhere.......
Current Issue
Special Report: Computing's New Normal, a Dark Reading Perspective
This special report examines how IT security organizations have adapted to the "new normal" of computing and what the long-term effects will be. Read it and get a unique set of perspectives on issues ranging from new threats & vulnerabilities as a result of remote working to how enterprise security strategy will be affected long term.
Flash Poll
The Changing Face of Threat Intelligence
The Changing Face of Threat Intelligence
This special report takes a look at how enterprises are using threat intelligence, as well as emerging best practices for integrating threat intel into security operations and incident response. Download it today!
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2020-14483
PUBLISHED: 2020-08-13
A timeout during a TLS handshake can result in the connection failing to terminate. This can result in a Niagara thread hanging and requires a manual restart of Niagara (Versions 4.6.96.28, 4.7.109.20, 4.7.110.32, 4.8.0.110) and Niagara Enterprise Security (Versions 2.4.31, 2.4.45, 4.8.0.35) to corr...
CVE-2020-11733
PUBLISHED: 2020-08-13
An issue was discovered on Spirent TestCenter and Avalanche appliance admin interface firmware. An attacker, who already has access to an SSH restricted shell, can achieve root access via shell metacharacters. The attacker can then, for example, read sensitive files such as appliance admin configura...
CVE-2020-13281
PUBLISHED: 2020-08-13
For GitLab before 13.0.12, 13.1.6, 13.2.3 a denial of service exists in the project import feature
CVE-2020-13286
PUBLISHED: 2020-08-13
For GitLab before 13.0.12, 13.1.6, 13.2.3 user controlled git configuration settings can be modified to result in Server Side Request Forgery.
CVE-2020-15925
PUBLISHED: 2020-08-13
A SQL injection vulnerability at a tpf URI in Loway QueueMetrics before 19.10.21 allows remote authenticated attackers to execute arbitrary SQL commands via the TPF_XPAR1 parameter.