hadoop - generating unique ids in hive -

i have been trying generate unique ids each row of table (30 million+ rows).

• using sequential numbers not not work due parallel nature of hadoop.
• the built in udfs rand() , hash(rand(),unixtime()) seem generate collisions.

there has simple way generate row ids, , wondering of has solution.

• my next step creating java map reduce job generate real hash string secure random + host ip + current time seed. figure i'd ask here before doing ;)

use reflect udf generate uuids.

reflect("java.util.uuid", "randomuuid")