hive-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ashish Thusoo <athu...@fb.com>
Subject Re: Efficient mechanism to simulate the row level updates in Hive
Date Wed, 16 Feb 2011 19:26:37 GMT
This is quite difficult to do in Hive on Hadoop. Hive over Hadoop really does not support row
level updates so basically you are reduced to periodically merging the raw stream of updates
with the main table and generating a new snapshot of the table. Another possible approach
could be to use hbase and run your updates to it and run Hive over hbase so that you can still
do adhoc querying on that data.

Ashish

On Feb 15, 2011, at 7:16 PM, Sheetal Dolas wrote:

Hello,

We have thousands of tables in a Hive database. Many tables have billions of records and multi
TB of data data in them.

We are looking for efficient mechanism to achieve row level updates on these tables.

Please share your experiences, ideas.

Thanks and Regards,
Sheetal




Mime
View raw message