Spearheaded by eBay, Fujitsu, and Verizon, the new group has four broad goals, the first of which is to establish the best practices for big data security and privacy, CSA representatives told InformationWeek in a conference call.
The second goal is to help industry and governments adopt these practices, said Big Data Working Group member Arnab Roy, a research staff member with Fujitsu Laboratories of America. The third is to establish relations with standards-development organizations to influence big data security and privacy standards. And the fourth is to "accelerate the adoption of novel research aimed to address security and privacy issues," Roy said.
"There are fundamental ways in which big data science differs from current technology," said Roy. One key difference is that big data often is collected from "diverse end points," including social networks and mobile devices, and might also include vast numbers of data owners, providers, and customers.
This information must be "aggregated and disseminated securely, and inside the context of a formal, understandable framework"--one that's part of a contract with the data owners, Roy said.
[ Read 10 Big Predictions About Big Data. ]
Another issue the group plans to address is data filtering, particularly as big data management systems ingest vast amounts of information, much of which won't be used.
"The balance between privacy and utility also needs to be thoroughly analyzed," said Roy. Organizations collect data to analyze, and hopefully glean, actionable intelligence from it, but they must also consider their customers' privacy needs.
The separation between data owners, providers, and consumers is another point of concern. "The integrity of the data coming from end points has to be insured," said Roy. In other words, organizations must find ways to prevent data poisoning by outside parties.
Finally, the group plans to address this big data issue: If security is compromised, how quickly can cloud providers migrate customers to another site?
The group plans to provide research and guidance on six themes, including big data-scale crypto, cloud infrastructure, data analytics for security, framework and taxonomy, policy and governance, and privacy. "We're going to try to set a baseline of best practices that we'll make available to the industry by the end of this year," said J.R. Santos, CSA's global research director.
By the end of 2012, the group plans to seek funding for its industry and government initiatives.
"Hopefully by Q1 of 2013 we'll start establishing more leads on relationships with a number of different standards development organizations," Santos said. "We'll look at things like security and privacy test beds, enabling sub-groups to execute on some of their plans, and also work with funding agencies."
The Big Data Working Group currently has more than 30 members, including representatives from industry and government agencies. It's chaired by Sreeranga Rajan, director of software systems innovation at Fujitsu Laboratories of America, and co-chaired by Neel Sundaresan, senior director and head of eBay Research Labs, and Verizon security expert Wilco Van Ginkel.
The CSA's decision to launch the Working Group was spurred by tech industry concerns around big data security and privacy. "Corporations have been approaching me directly about wanting to do this," said Santos. "Fujitsu was one of those companies. It had talked about the research they were doing. This is one of the bigger issues in the cloud, and we needed to move forward with it."
High-frequency trading (HFT) has been in the spotlight recently as market participants have debated the value of HFT for investors. While only a certain number of players compete at the ultra-low latency level, many other players are taking advantage of technologies utilizing models and strategies that leverage new data sets as well as more types of information. Join Wall Street & Technology for the HFT Under Attack webcast to find out the opportunities of using big data in a trading strategy. It happens Sept. 12.
