Ulf Mattsson: We’ve seen studies talking about how Big Data analytics can revolutionize the way the Internet did many years ago. I’ve seen figures saying that 63% are using it to enhance customer relationships, 58% are using these capabilities to redefine the product and product development, and 56% are using it to change their operations. These create a lot of opportunities but, at the same time, open up privacy issues. According to recent reports and studies, 51% are saying that security is the most important issue. The monetization of data and data breaches are driving the need for security.
Most of Big Data is based on Hadoop and Hadoop is introducing new system layers for security. Traditional security approaches are not working because they are typically based on some kind of control in-boxing or choke point. You need to know where the data is. If you box it in and build walls around the it, Big Data will not work because Big Data is based on data moving around between the different nodes. It’s very high performance so you cannot use your traditional parameters of security. So, you to address not just the issue of Big Data where you have the new security approach but also the data flow.
What is recognized as a big issue recently by Gartner is that we actually have different silos in the enterprise. We have different islands of legacy systems that are feeding data into Big Data. You have enterprise warehouses that are using the data from Big Data. It’s again about the data flow. Gartner is pointing to breaching the silos as a cryptic way to solve the security problem going forward. When they talk about breaching these silos, they are referring to an enterprise policy. Your security policy would, on an enterprise level, control it from a single point instead of point solutions for each silo.
These enterprise policies should then be automatically enforced on these different silos and platforms. We use the term “bridging the gap” for this where you can actually be compliant and protect privacy, and at the same time use the data. With traditional security approaches, you tend to lock down the data and then the data will not be available for analysis. You need to use another approach.
What Gartner is also acknowledging is that you need to think about the granularity. If you have a more granular way to protect different data and different feeds, then you can adjust the protection. You get the balance between security and utility. You actually can use your data for analytics and still be compliant and have control over the protection.