February 24, 2016

Security Concerns Grow As Big Data Moves to Cloud

George Leopold
red hacked symbol

Despite exponential increases in data storage in the cloud along with databases and the emerging Internet of Things (IoT), IT security executives remain worried about security breaches as well as vulnerabilities introduced via shared infrastructure.

A cloud security survey released Wednesday (Feb. 24) by enterprise data security vendor Vormetric and 451 Research found that 85 percent of respondents use sensitive data stored in the cloud, up from 54 percent last year. Meanwhile, half of those surveyed said they are using sensitive data within big data deployments, up from 31 percent last year. One-third of respondents said they are accessing sensitive data via IoT deployments.

The upshot is that well over half of those IT executive surveyed are worried about data security as cloud usage grows, citing the possibility of attacks on service providers, exposure to vulnerabilities on shared public cloud infrastructure and a lack of control over where data is stored.

Those fears are well founded, the security survey notes: “To a large extent both security vendors and enterprises are like generals fighting the last war. While the storm of data breaches continues to crest, many remain focused on traditional defenses like network and endpoint security that are clearly no longer sufficient on their own to respond to new security challenges.”

Control and management of encryption keys is widely seen as critical to securing data stored in the cloud, the survey found. IT executives were divided on the question of managing encryption keys, with roughly half previously saying that keys should be managed by cloud service providers. That view has shifted in the past year, the survey found, with 65 percent now favoring on-premise management of encryption keys.

In response to security concerns, public cloud vendors like Amazon Web Services, Google, Microsoft and Salesforce have moved to tighten data security through internal development, partnerships and acquisitions in an attempt to reduce vulnerabilities. Big data vendors have lagged behind, but the survey noted that acquisitions by Cloudera and Hortonworks represent concrete steps toward securing big data.

Cloudera acquired encryption and key management developer Gazzang in 2014 to boost Hadoop security. Among Hortonworks’ recent acquisitions is XA Secure, a developer of security tools for Hadoop.

Still, the survey warned, IoT security remains problematic.

When asked which data resources were most at risk, 54 percent of respondents to the Vormetric survey cited databases while 41 percent said file servers. Indeed, when linked to the open Internet, these machines can be exposed vulnerabilities similar to recent “man-in-the-middle” attacks on an open source library.

(Security specialist SentinelOne released an endpoint platform this week designed to protect enterprise datacenters and cloud providers from emerging threats that target Linux servers.)

Meanwhile, the top security concerns for big data implementations were: the security of reports that include sensitive information; sensitive data spread across big data deployments; and privacy violations related to data originating in multiple countries. Privacy worries have been complications by delays in replacing a 15-year-old “safe harbor” agreement struck down last year that governed trans-Atlantic data transfers. A proposed E.U.-U.S. Privacy Shield deal has yet to be implemented.

Despite these uncertainties and continuing security worries, respondents said they would continue shifting more sensitive data to the cloud, databases and IoT implementations as they move computing resources closer to data. For example, half of all survey respondents said they would store sensitive information in big data environments.

Recent items:

Pulling the Threads on Big Data Security

A Peek Inside Cisco’s Hadoop Security Machine

Share This