Hi Raajay,
Regarding how hive build DAG, you can check
https://github.com/apache/hive/blob/72f97fc7760134465333983fc40766e9e864e643/ql/src/java/org/apache/hadoop/hive/ql/exec/tez/TezTask.java
Check method TezTask#build
Do you mind share what kind of changes you want to make on DAGs ?
On Tue, Se
Is there any way to see the operator tree of hive before and after
optimization ? Thanks
I mean if I want to run on hadoop 2.x, do I must to build it from source
with hadoop 2.x ?
Hive manual says that hive don't support "no two aggregations can have
different DISTINCT columns". I think theoretically it is possible to
implement this, is there any specific reason for that ? Performance reason
or something else ?
I use the following sql with mr engine and find that it would invoke 3 mr
jobs. But as my understanding the join and group by operator could be done
in the same mr job since they are using the same key. So not sure why here
still 3 mr jobs, anyone know that ? Thanks
select s2.name,count(1) as cn
