Skip the navigation

Big data can be a big headache for data defenders

By John P. Mello
April 25, 2013 09:30 AM ET

CSO - Big data is all the buzz in board rooms these days, but in security circles it's a mixed blessing.

The technology allows tons of threat data to be analyzed and used to thwart intruders. In the enterprise, it can be used to analyze volumes of network traffic to reveal insider trouble and advanced persistent threats.

Problem is, the network architecture that supports big data wasn't created with security in mind, which can create problems for system defenders.

A common big data deployment uses Hadoop to manage the clusters of computers used to crunch the mountains of data fed into it. That can create problems for defenders, especially if they're trying to use traditional security tools to protect their big data fortress.

"Incumbent data security vendors believe that Hadoop and distributed cluster security can be addressed with traditional perimeter security solutions such as firewalls and intrusion detection/prevention technologies," explained Zettaset, a big data security company, in a white paper "The big data Security Gap: Protecting the Hadoop Cluster."

"But no matter how advanced, traditional approaches that rely on perimeter security are unable to adequately secure Hadoop clusters and distributed file systems," it added.

Traditional security products are designed to protect a single database, noted Zettaset CTO Brian Christian. When those products are forced to protect a distributed cluster of computers that can number in the thousands, their capacity is taxed.

"When you put them on a large scale distributed computing environment, they become either a choke point or a single point of failure for the entire cluster," Christian said in an interview.

"They could potentially be extremely dangerous running them on a cluster," he said, "because if they do fail, there is the potential to deny everybody on the cluster access to petabytes of data or a corruption of data in some of the encryption security technologies."

[Also see: RSA 2012: The security risks and rewards of big data]

Performance problems aren't unusual where security is "bolted on" to a big data setup, said Jason Escaravage, a principal with Booz Allen Hamilton.

For example, if you have a central place where everyone has to check credentials and you have to leave the system to access those credentials on an external security system, each one of those transactions are going to slow things down.

That can be challenging not only for system performance, but for a company's bottom line, too. "When you come to a system that's already been deployed and you're trying wrap security around it or bolt it on at the end, it's costly because it wasn't designed up front," Escaravage said in an interview.

Originally published on www.csoonline.com. Click here to read the original story.
This story is reprinted from CSO Online.com, an online resource for information executives. Story Copyright CXO Media Inc., 2006. All rights reserved.
Our Commenting Policies
Internet of Things: Get the latest!
Internet of Things

Our new bimonthly Internet of Things newsletter helps you keep pace with the rapidly evolving technologies, trends and developments related to the IoT. Subscribe now and stay up to date!