Evaluating Hadoop

Brett Sheppard at O’Reilly has an excellent, lengthy article aimed at helping organizations considering Hadoop, covering what to consider when picking a distribution to setting up your initial production cluster. Sheppard starts off by giving us a brief history of Hadoop, which really has come a long way in a short period of time. The article contains everything you need to begin an initial evaluation in pseudo-distributed mode, as well as bases for selecting an appropriate installation for your organization. He gives an excellent overview of several different installations and then walks you through subsequent steps such as planning your architecture.