Commonly Used Hive Setting

I just list very commonly used ones.

Set dynamic partition (a column name)

SET hive.exec.dynamic.partition=true;

Affect insert rows to save in sampled format

SET hive.enforce.bucketing = true;

Reduce output compression

SET hive.exec.compress.output=true;

Map output compression

SET hive.exec.compress.intermediate = true;

Set codec which need to be installed

SET mapred.output.compression.codec = 
 org.apache.hadoop.io.compress.SnappyCodec;

Enable Hive’s automatic join optimization to convert repartition joins to replicated joins (map side join only) if one of the tables is small enough.

SET hive.auto.convert.join = true;

Below two tells Hive to optimize joins where it sees skewed data. Sets the threshold beyond which a key is considered to be skewed. For skew data, separate join will be created to handle them.

SET hive.optimize.skewjoin = true; 
 SET hive.skewjoin.key = 100000;