I have very generic question but I am looking to get some guidance
from the community.  I have a  proprietary database system with
SQL interface. But, we are exploring the opportunity to submit the SQL by
the client and my query can go to hadoop or the proprietary database.
       My question is: what are the options for me to do it and their pros
and cons.
I kmnow few options like: using odbc driver, using rest api or using thrift
Does anyone has experience with them. It would be great to hear their
ur lops's gravatar image asked Apr 10 2013 at 17:47 in Hive-User by ur lops

1 Answers

hello Rob,
hive does have both jdbc and hiverserver as well so you can connect them remotely. the issue here would be more of 1) how will you keep the data in sync between hive and your datastore 2) There are plenty of queries which work on properietary databases but will not work on hive as it is 3) how are you planning to decide which query goes to which datastore?
Others will have far more details questions as well but never heard anyone doing a dual live setup where queries can run anywhere on traditional data stores as well as hive on the fly
Nitin Pawar's gravatar image answered Apr 10 2013 at 18:09 by Nitin Pawar