At the beginning of a recent customer meeting I was involved in, I mentioned how easy it is to shut down idea generation by judging too soon or too harshly, to which a CIO at the meeting shared this story about five monkeys in a cage. You may have heard it before:

Begin an experiment by placing five monkeys in a cage and some bananas at the top of the cage. All five monkeys will attempt to climb the cage to reach the bananas, but if you have someone knock them down with a fire hose every time they get close, eventually the monkeys will stop trying. Now replace two monkeys with two new monkeys and these two new monkeys will attempt to climb up and get the bananas; however, the original monkeys will try to "help" and pull the new monkeys down to prevent them from being hit by the fire hose. Once again all five monkeys will eventually give up climbing for the bananas. Replace another two monkeys and so on, eventually you will have five monkeys in the cage who will not try to reach the bananas and none of them even knows why they don't try.

No, I am not calling anyone a monkey, but this story does highlight how easy it is to prevent innovation from occurring simply by not fostering an environment that encourages open discussions.

How does this relate to SAS High Performance Analytics? I challenge you to think about problems your company has attempted to answer in the past but failed to solve because of self-imposed constraints that prevented success. This thinking about self-imposed constraints resonated so well with another customer that I met with last week that it caused a valuable brainstorming session to occur as we discussed serveral business problems they had failed to address in the past due to IT-imposed processing constraints.

Here are a couple of examples of what self-imposed constraints may look like within an organization:

• IT imposed batch windows that limit the amount of data that can be processed in a timely manner (timely manner using traditional compute environments, meaning non-high-performance analytic environments).
• Sampling large data to analyze ONLY because it wasn't possible to do analysis on that "large" set of data (in a non-high-performance analytics environment).
• Not attempting to analyze a problem because it was attempted in the past and the size of the data simply made it impossible to compute an answer, especially in a timely manner (in a non-high-peformance analytics environment).

What other types of self-imposed constraints can you think of that may be holding you and your organization back from solving an important issue? How might you solve them once you remove the constraints with a high-performance analytics environment?

Learn more from big data experts in this special 32-page report on high-performance analytics.