|By Joseph Parker||
|December 21, 2013 04:00 PM EST||
Big Data is bandied around so much that it seems to have lost what little meaning it had in the first place. What's so special about it, anyway? Some people think it's just a lot of data, more than most companies are used to (Google doesn't count). The real definition isn't much more … well, definite. It is, as Tech Target's glossary says, "… the voluminous amount of unstructured and semi-structured data a company creates — data that would take too much time and cost too much money to load into a relational database for analysis."
How much is "too much"? That's the tricky part; it largely depends on how you set up your database(s) and the hardware you use. For practicality, we'll just say that when you move your siloed data to a non-relational database, you're using big data. Many businesses are switching to a big-data capable infrastructure, enough that Gartner Research predicts that this switch is going to drive $232 billion in spending through 2016. Why are companies spending so much on a single area?
Though your company likely has enough types and volume of data to earn the moniker "big," you can't just throw it all in a single database and expect to see quick, reliable benefits. Hence, the hundreds of billions of dollars companies will spend in the next few years just to support these massive non-relational databases. You can't dump your data onto a traditional hard disk NAS and run reports in the time it takes to get yourself a cup of tea. Hard drives are notorious for poor random I/O performance, so your big data solution needs a solid flash NAS or SAN array that can access non-sequential data quickly.
You'll have to beef up your servers, too. Running a report on a relational sales database is easy for most servers when compared to running a sales report on a big database. Make every member of your organization's life easier by running your database in a cloud server — public, private, in-house, outsourced, hybrid, whatever type is right for you.
But why is big data so important? What makes it attractive to so many companies? Again, we'll turn to Tech Target, though I won't quote it directly:
· Patterns: Whether you're a B2B shipping container manufacturer or a B2C online retailer, you need more information about your business, your customers and your industry. The more data you collect in one place, the easier it is to see patterns. What's your busiest time of year? Your bestselling product/service? Are most of your customers from the same region? Are people more likely to buy from you after reading 12 or more pages on your website? Having that data available and waiting for you to dig into is invaluable.
· Space: How much duplicate data is your business storing that you could eliminate by combining all your disparate databases? Are you perhaps collecting too much data? I know my previous point harped on collecting more data, but if you're storing information that serves no purpose, it's a lot easier to eliminate if it's all in a single location.
· Legal Compliance: If you don't have a complete view of the data you retain, it's hard to make sure you're complying with all applicable laws and policies. If someone sues your company or you undergo a compliance audit, you need to have easy access to all the information you're storing.
It's easy to dismiss big data as some overhyped phrase people throw around, but when you dive deeper, you realize that it serves a practical purpose. The days of every department having its own siloed database aren't over yet, but they should be. The advantages of big data ensure that.