What are pros and cons of using expansionary and contractionary fiscal and monetary policy tools?

What are the pros and cons for using HDFS vs. Gluster (especially for Hadoop)?

• We're planning a medium-size farm for Hadoop tools.

• Answer:

  HDFS is (of course) the filesystem that's co-developed with the rest of the Hadoop ecosystem, so it's the one that other Hadoop developers are familiar with and tune for.  It's also optimized for workloads that are typical in Hadoop. GlusterFS is a general-purpose distributed filesystem.  If you want to be able to access data both within and outside of Hadoop, without having to manually migrate data in or out of the Hadoop universe, then that's an advantage.  GlusterFS is also fully/properly distributed, so it doesn't have a single point of failure like the HDFS NameNode. The biggest disadvantage to using GlusterFS with Hadoop currently is that Hadoop works best when it can locate jobs near data.  This functionality is fully implemented and regularly used with HDFS.  The same functionality (allowing map/reduce to query storage about locations for specific data) has been implemented for GlusterFS as well, but won't appear until 3.3 and requires some very minor configuration before map/reduce can use it.