Implementing and Optimizing the Transition

Abstract

Recently, I have seen a number of organizations reversing their decision about using Big Data. These organizations represent diverse industries and therefore may not withstand the logic of Hadoop not being suitable for a particular industry. In most of these cases (apparently), performance and lack of usability were the main issues for deciding against using Big Data technologies. Granted that open source software does not have extensive and easy-to-understand documentation or means to troubleshoot, but these days, there are enough vendors selling supported Hadoop distributions and software with easy-to-use browser interfaces for performing all kinds of tasks starting with data ingestion to the end result: analytics.