Big data has taken the tech industry by storm. Tips, tricks and how-to guides are everywhere you look. According to IT experts at the Enterprise Strategy Group, 47 percent of enterprises are now collecting and analyzing more than six terabytes of security data on a monthly basis. The increase in popularity has led way to quite a lot of hype, so how can you tell fact from fiction? Let’s explore four common big data myths.
Myth #1: “My current IT solutions will suffice”
No, they will not. To put things in perspective, if every organization’s existing security solution were sufficient, would the number of cyberattacks continue to grow? Would the cost of these attacks continue to increase? Would the time and cost of resolving threats continue to climb? Stored in data warehouses and platforms, data is continuing to grow exponentially. If our current IT solutions were sufficient, we would not hear about the risks we face, and we would be able to make use of all the data. However, this is where the challenge lies: How do you make use of all the data? Is the information solving any of your organization’s problems or challenges? The truth is that the answers are in the data, but advanced analytics are required to sort through and identify them.
If you map it back to Gartner’s Hype Cycle, big data is at the trough of disillusionment. Why? Because organizations are not yet equipped with the technology to discover what purpose the data can serve. Organizations need to figure out what to do with the data they have, and what applications they should use to analyze and manage it. Big data alone does not provide organizations with the answers they need to create a more efficient, secure and profitable business. The answer is to use data analytics AND visualization. With data analytics and the right visualization framework, organizations can create an interactive, intuitive and dynamic means for data analysts to make smarter decisions.
With the proliferation of applications, devices and an increased volume of data, it’s now more difficult than ever to protect against cyberthreats. Big data analytics and visualization are the new frontier.
Myth #2: “Big data demands data scientists, who are expensive and hard to come by”
Big data does not demand more data scientists. The volume of data is growing exponentially, so it’s unrealistic to expect security professionals to analyze vast quantities of data in a short time period. Even with the right skills and experience, manpower will always be dwarfed by the information we’re collecting – and people are expensive. Merely hiring more and more data analysts and data scientists leaves room for error and inefficiency.
Organizations should rely on technology to produce intuitive and easy-to-digest visualizations that pinpoint the correct data set. Using such tools, data scientists spend less time filtering through unimportant data and more time identifying solutions that provide actionable insights quickly and efficiently.
Myth #3: “You need all of the data”
In an ideal world, organizations could leverage all the data they collect. Storing that information can be costly and time consuming. The greater the volume of information, the longer it takes to analyze, which means you’ll need powerful and expensive hardware capable of mass processing. You should ask, then whether the level of accuracy significantly increases when you collect and store all of it, or could you have received the same answer by storing and looking at only a sample of the data?
The right technology partner will understand what data should be captured to get the results you need,
Myth #4: “Big data just means Hadoop”
Sure, if all you wanted to do is store the data, big data could be synonymous with Hadoop. But wait, isn’t the problem you’re having extracting insight from big data? If you answered yes, then big data in no way means Hadoop only. In a recent webinar, Jon Oltsik senior principal analyst with ESG confirmed this idea. Hadoop is likely one piece of your big data solution, but it is not necessarily a solution that is able to address all types of insights. While Hadoop is a great platform to address big data issues, it is imperative to select a platform specific to the type of analytics you want to run on a particular big data set.
While these do not nearly cover all of the myths out there, they highlight some of the most concerning ones.
John Trobough is the president and CEO of Narus (News - Alert). John brings more than 20 years of operations and international experience in the telecommunications and mobile software industries to the company.