Tuesday, October 21, 2014

Big Data, Big Puzzle

Big Data is about harnessing largesse and being adaptive enough to respond to change.

Big Data is still a big puzzle for many organizations. The term "Big Data" was first used as a reference for all the data generated by/through the internet...the largest amount of data ever generated by mankind... afterward. "Big Data" is a concept where at its purist and fundamental level is the attempt to try to leverage the inner-connected nature of data that is collected during the operation of any enterprise. 

So in a nutshell, "BIG DATA" means crap lots of data -it is data that needs more than just the standard commodity products, databases and systems to support it. If it can be supported by commodity products (traditional load balancing engines, etl tools, shared-everything SMP servers and standard network devices), and if it can be said that the query turnaround time doesn't really matter, then the enterprise has no real "big data" issue. Their data's not big enough to qualify and their data demands are not pressing enough to move-the-needle. On the other hand if an enterprise has data tables that are in the tens-of-billions of rows, of structured data, then it is very difficult to disqualify this as "big data". It's data and it's pretty big.

Five “V”s of Big Data: Volume, Velocity, Variety, Veracity, and Value. Big Data can mean large data sets but is more about the combination of different types of data together that allow you to answer key questions about your business. It can be as big and complex as you want it to be but needs to answer the critical questions you can't answer by looking at multiple systems with tidbits of data. Big Data is nothing more nothing less than data with more Volume, at insane Velocity, coming from a variety of sources and with high Variety when it comes to the metadata. It also has to be true (Veracity) and it has to bring value to your working environment

Big Data initiative: Many organizations are taking Big Data initiatives, such initiatives may have more than one of the following characteristics:
- Petabytes of data used for analytics
- denormalized / semi-structured and unstructured data used for analytics
- the implementation of Hadoop and related technology
- the use of analytic data for predictive analyses and data mining

Dig through the HOW: Big Data is not being utilized to its full potential for anything and everything from analytic to process. After clarifying the WHAT is Big Data, let's "start focusing on the HOW ".
- How many types of Data you are gathering / generating/creating / capturing as our Inputs for our Organization? 
-Whether the present Inputs (if any) are sufficient for the Organizational Goals / Aims? 
-Whether any duplication s found in the present Inputs? Can it erected & shorten the Inputs? 
-What & where you should add new Points / information in previous Inputs Formats to get the proper results? 

The "big data" realm is not about unstructured data, but about harnessing largesse and being adaptive enough to respond to change. To put simply, Big Data is not the end, the end is how to capture the business insight from the big data, and leverage it for enabling business growth and improving customer satisfaction.


Post a Comment