Big Data Demands Big Computing - InformationWeek

InformationWeek is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Data Management // Big Data Analytics
09:36 AM
Keith Collins
Keith Collins
Connect Directly

Big Data Demands Big Computing

Big data has been around a long time. Here are four best practices to help you tap into the big computing power that is finally unlocking value.

10 Cloud Analytics & BI Platforms For Business
10 Cloud Analytics & BI Platforms For Business
(Click image for larger view and slideshow.)

The world has gone gaga for big data. Yet, companies still struggle to understand how collecting and analyzing data can provide a deeper view of customers and lead to better business decisions.

In the midst of all the big data hype, we seem doomed to the same fate encountered in the enterprise data warehouse (EDW) era, in that we don't understand what has changed. You may have noticed that the question often asked today is the same question that was asked when the EDW came into vogue: "How big is it?"

The question should be: "What can we learn from this data?"

[ Want more on this topic? Read Gartner Advanced Analytics Quadrant 2015: Gainers, Losers. ]

We haven't grasped that computational power -- or big computing -- is the real change that has opened the door to big opportunity. Big computing at small prices allows companies to look at, and deal with, data in ways not possible before. It's this computational capacity that has the real potential to transform data from a compliance burden into a business asset.

Organizations have always collected data, but until recently, large-scale cluster computing and analytic algorithms that could perform at scale were cost-prohibitive. That's no longer the case, and many organizations are now experimenting with big data. But they're not investing in skills and tools to analyze that data, a situation akin to planting a garden and then not watering it.

In 2014, big data was defined by masses of unstructured content. This year, big data will be defined increasingly by sensor data captured from the Internet of Things. Leaders in each industry are beginning to find real value in unlocking the potential in the data they already have. The insatiable desire to visualize data, recognize patterns, and turn data into dollars is being supercharged by high-performance computing.

In the latest wrinkle, the conversation is shifting from big data to machine learning. The foundations of machine learning were established in the 1950s. It now sits at the intersection of disciplines including artificial intelligence, statistics, and data mining. Leading work came from artificial intelligence pioneers like Alan Turing, but the concepts weren't widely adopted until the 1990s. Why? For one thing, the computations were too expensive for mass adoption at the time. Today, we can run billions of simulations to teach a machine to play poker or learn the concepts that define a cat.

The principles of statistics, forecasting, and optimization remain the same, but to seize the opportunities in big computing, it's imperative to modernize your playbook in four ways:

Offer education. Invest in educating the IT teams that support data scientists, from business analysts to database architects. These people need to understand the basics of analytics in order to appreciate both the art of the possible and the impact of organizing data for the task of rigorous analysis.

Ensure data access. Focus on data governance to accelerate and improve access to data, not to restrict it. Speed and agility require easy, yet managed, access.

Enable exploration. Invest in analytic "sandbox" environments that support large-scale cluster computing. Provide a toolbox for your data scientists, along with the computing capacity to look deep into your data, so they can find hidden relationships and meaning.

Be agile. Let the explorers fail fast, but also be sure to market their successes. The returns from combining big computing and analytics are like compounding interest: It's a gift that keeps on giving.

Attend Interop Las Vegas, the leading independent technology conference and expo series designed to inspire, inform, and connect the world's IT community. In 2015, look for all new programs, networking opportunities, and classes that will help you set your organization’s IT action plan. It happens April 27 to May 1. Register with Discount Code MPOIWK for $200 off Total Access & Conference Passes.

Keith Collins is the CIO at SAS. View Full Bio
We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.
Comment  | 
Print  | 
More Insights
Newest First  |  Oldest First  |  Threaded View
User Rank: Apprentice
10/13/2015 | 3:29:39 PM
We all know big compute and big data or dark data has been around for a long time. This is not the change that is making the real time digital democracy happen. We all know that EDW Evolution was not perfect, however it was a great way to get to a sort of standard in and around mainframe best practices. What is happening now is that through the use of compute access and API Portfolio's data is being managed more openly and used in a productive way with crisp and efficient approaches up and down the IT Enterprise. These approaches have 4 key areas of consideration. Connect, Protect, Create and Clafrify data, insights and services to a given user or professional organization. These data approaches tend to be extensions of the existing EDW model and Cloud services rather than use discrete private math and data management as they fit right in line with existing investments and the IT Professional base who runs the internet based business process. IT Professionals. There 25 million IT PROS on planet earth we will be at the center of these new data business needs and we are being trained to automate the role of the data scientist, data officer. IBM, ORACLE, MS, DELL get this as do SALESFORCE, GOOGLE.. Which begs the question... How does SAS Play in this field with no consitent relationship or cultural acceptance of IT Professionals ? QED...
User Rank: Ninja
3/8/2015 | 1:16:36 PM
One step closer to IoT
Big data is making great gains as more firms develop stronger maturity levels needed to gain and ultimately apply insights. Of course the next evolution is to embrace IoT as a means to bring more data into the fold and empower the machine learning you discussed at more granular and meaningful levels. Understandably as this progression continues we also need to pay very close attention to how we are handling security. As we enter into an IoT environment, the security approach commonly utilized today will prove inadequate. It's time for a true evolution in how we view security. Peter Fretty, IDG blogger working on behalf of Cisco. 
D. Henschen
D. Henschen,
User Rank: Author
3/2/2015 | 11:15:01 AM
Cheap computing has made the difference
Here's an important reminder that big data has been around for a while. It's low-cost, distributed computing that has made the difference in the so-called big data era. Big data platforms like Hadoop and NoSQL databases (and data warehouse appliances before them) unlocked the power of many low-cost, industry-standard servers.
2021 Outlook: Tackling Cloud Transformation Choices
Joao-Pierre S. Ruth, Senior Writer,  1/4/2021
Enterprise IT Leaders Face Two Paths to AI
Jessica Davis, Senior Editor, Enterprise Apps,  12/23/2020
10 IT Trends to Watch for in 2021
Cynthia Harvey, Freelance Journalist, InformationWeek,  12/22/2020
White Papers
Register for InformationWeek Newsletters
Current Issue
2021 Top Enterprise IT Trends
We've identified the key trends that are poised to impact the IT landscape in 2021. Find out why they're important and how they will affect you.
Flash Poll