next environment configuration is as follows:
CentOS 6.7
Hadoop2.7
Hive 2.0
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Want to be used in the machine hive on the spark, which hadoop2.7 support spark version is 2.0, see the introduction of the website ttps://cwiki.apache.org/confluence/display/Hive/Hive+on+Spark:+Getting+Started, needs to be compiled not including hive jar spark, so their download the source code to compile, but spark2.0 source and previous versions have the difference, not the make - distribution. Sh file, can only use SBT or build maven, but want to realize how with hybrid function, consult everybody,
