What Does Taming Big Data Really Cost?
By Bill Franks, Sep 11, 2013
I’ve written in the past about the hype that surrounds big data. Perhaps one of the most ridiculous urban legends out there today is that working with big data is cheap. This myth is largely driven by the fact that some big data tools, such as Hadoop, don’t come with a license fee.
Before I get started, let me be clear on one thing. Open source products such as Hadoop have many uses and can play important roles in the development of analytic processes. Plus, us stats geeks love to download useful software for free just like anyone else. Who doesn’t want free stuff? The key is to look at what it will cost in total to actually use that software in the context in which you plan to use it. If you aren’t careful to take into account the entire spectrum of costs for your analytic platform, you might just find yourself going way over your planned budget.
Many Price Is Right winners have learned the hard way that their “free” RV came with a large tax bill and heavy ongoing maintenance and operational costs. If you can’t handle those costs, then winning an RV isn’t as exciting as it seems. Similarly, getting a free puppy leads to a lot of ongoing work, as well as costs for food, vets, and myriad other items. It isn’t that winning an RV, or getting a puppy, or deciding to leverage Hadoop is a bad thing. It is simply a matter of being sure you understand what you’re committing to.
Look Beyond License Fees
Regardless of what you pay for your software, license fees are but the beginning of the overall cost of operation for an analytic platform. Other costs include:
- The hardware that the software will be installed on
- The space taken and power used by the hardware
- Configuring & implementing security, resource prioritization, and other operational features
- Acquiring, loading, and making data ready for use
- Developing analytic processes on your platform
- The latency between requests and results being delivered
- Maintaining the platform
- Training staff to use and configure the platform
- Consultants needed for any part of implementation
Certainly there are other costs, but the point should be clear. You can’t simply look at any one factor when determining your costs and deciding which direction to go. A simple comparison of license fees or cost per server won’t lead you to the right decision. In addition, it is also important to look at total cost over the expected life of the platform. Saving on license fees can lower first year costs, but if ongoing costs are more expensive, that can be more than made up for over time.
I believe that the most often underestimated line items relate to the man hours required from employees or consultants to actually stand up, configure, utilize, and maintain an environment. It is absolutely critical to account for these costs. This is one area, for example, that Hadoop can run up your bill. It is a newer, maturing technology that few people are yet familiar with. This leads to a steep learning curve. Again, this isn’t to say the learning curve can’t be worth it. I am simply saying that you have to recognize and account for it.
One of the best discussions I have seen on this topic is a recent paper from Richard Winter at Winter Corp. I encourage you to download and read his report. In it, he develops a framework and provides some examples that lead to very different conclusions on where to invest based on the profile of the data being targeted and the analytics required. He shows how in some cases, Hadoop is a solid fit and in others it doesn’t fit well at all. That’s also true with other technologies. As always, it comes down to what you need to do.
As Teradata states with its Unified Data Architecture, there is a place for Hadoop and other open source products. There is also a place for commercial technologies such as Teradata. Each has a unique total cost profile based upon the type and volume of processing required. By focusing on your total costs, you’ll be better able to allocate your resources over time and you’ll ensure that the right tools and technologies are being used in the right ways.
If you’re reading this, your action item is to make sure your organization is considering all the costs it will incur over time in order to make a solid, appropriate decision as to what tools and technologies serve what roles in your environment. Otherwise, you’ll make it harder for your organization to tame big data.
About the author
Bill Franks is Chief Analytics Officer for Teradata, where he provides insight on trends in the analytics and big data space and helps clients understand how Teradata and its analytic partners can support their efforts. His focus is to translate complex analytics into terms that business users can understand and work with organizations to implement their analytics effectively. His work has spanned many industries for companies ranging from Fortune 100 companies to small non-profits. Franks also helps determine Teradata’s strategies in the areas of analytics and big data.Franks is the author of the book Taming The Big Data Tidal Wave (John Wiley & Sons, Inc., April, 2012). In the book, he applies his two decades of experience working with clients on large-scale analytics initiatives to outline what it takes to succeed in today’s world of big data and analytics. The book made Tom Peter’s list of 2014 “Must Read” books and also the Top 10 Most Influential Translated Technology Books list from CSDN in China. Franks’ second book The Analytics Revolution (John Wiley & Sons, Inc., September, 2014) lays out how to move beyond using analytics to find important insights in data (both big and small) and into operationalizing those insights at scale to truly impact a business.He is a faculty member of the International Institute for Analytics, founded by leading analytics expert Tom Davenport, and an active speaker who has presented at dozens of events in recent years. His blog, Analytics Matters, addresses the transformation required to make analytics a core component of business decisions. Franks earned a Bachelor’s degree in Applied Statistics from Virginia Tech and a Master’s degree in Applied Statistics from North Carolina State University. More information is available here: .