The use of big data analytics has allowed feds to increase cybersecurity, but agencies still face challenges processing a substantial percentage of the collected data. This hampers the ability to reduce breaches, according to "Navigating the Cybersecurity Equation," a new report from MeriTalk and Cloudera.

The partnership's survey of 150 federal IT managers found a vast majority employing big data — 81 percent said big data analytics is used for cybersecurity in some capacity, 53 percent said it's within their overall cybersecurity strategy and 28 percent said it's used in a limited capacity. This is a significant increase from 2013. Ninety percent of those surveyed acknowledged big data analytics has contributed to a decrease in some aspect of cybersecurity, including malware, insider threats and social engineering.

Download: Navigating the Cybersecurity Equation

Ninety-four percent of participants plan to invest in facets of big data analytics, including technology of infrastructure, hardware and business intelligence tools. 

At the same time, 88 percent believed data is not processed efficiently enough because of a lack of skilled personnel, potential privacy concerns and a lack of management support, and it’s estimated around 40 percent of data is not utilized. Currently, 59 percent of those surveyed reported dealing with cybersecurity compromises at least once a month, and the ability to sift and sort data in a timely manner can improve the ability to identify and monitor vulnerabilities.

"Internal and external cybersecurity threats will continue to evolve daily and we need to unlock the power of the data in order to regain the advantage," said Rocky DeStefano, a cybersecurity expert with Cloudera. "Agencies need complete visibility into the data across their enterprise. These teams also need the ability to flexibly analyze that data in a meaningful time frame so they can detect advanced threats quickly, identify the impact and reduce the associated risk. Accelerating investment in the platforms necessary to collect and analyze this data is critical to the success of these programs."

Those surveyed include chief information officers and chief technology officers; security analysts, architects and consultants; software development and data center managers; and individuals from in both civilian and Defense Department/intel agencies. The report has a margin of error of ±7.97 percent at a 95 percent confidence level.