Mike's Maxims for Hadoop

  • Once you have a strategy for managing your data architecture, adding nodes will help the performance of the Hadoop in that architecture; if you have no strategy, adding nodes will let useless processes run really fast.
  • A Hadoop cluster needs some amount of systemic (planned) care and feeding to allow for normal expansion of capacity and occasional reallocating of services across nodes of the cluster to meet new demands as you achieve new usage thresholds.

Leave a Comment

Your email address will not be published. Required fields are marked *