Challenges in Hive Implementation
Challenges :
There are many real time problems where in we need to use nested queries , whereas hive supports only correlated queries.
Hive does not have subtraction operation  and thus we need to create two tables and perform left outer join on
 it with condition to accomplish the task.
  some predefined Teradata functions are not available in Hive 
 Some calendar related predefined tables are not available in Hive
 No restart capability so we need to use batch_id as extra column
 Union all there is a complication and as a work around we need to use Temp Table concepts
Hive Problems Vs Teradata :

Subqueries wont work Nested subqueries are not supported
We need to use multiple sqls in hive
Calendar views in metadata
Hive dont have non equi joins ( >,<) . we need to make Mapjoin (data to be stored in every node ).
Qualify keyword not available in HIve , we need to use derived table for the same
Set Operation not allowed

Challenges faced hive : 

Heap errors 
Not Like operators
Version Conflict of Hive version in AWS
Parent child relation ship is tough is sqoop