Here is a workaround to insert large *batch *of data, each batch insert is done on their own partition.
create table hdfs.`/mytable/1` as (select....) Later on, If you want to insert new batch of data to the same table: create table hdfs.`/mytable/2` as (select....) Then when you query select * from hdfs.`/mytable` will include results from partition 1 and 2. If you query while a partition is generated, error may occurs. As a workaround for this, you may use a script (we use python) that ctas a batch of data on a tmp folder then move the tmp table to the mytable location as a partition. Hope it helps On Thu, Nov 17, 2016 at 9:02 AM, Arina Yelchiyeva < [email protected]> wrote: > Hi Sanjiv, > > currently Drill doesn't support insert or update of data in tables. > > Kind regards > Arina > > On Thu, Nov 17, 2016 at 3:54 PM, Sanjiv Kumar <[email protected]> wrote: > > > Hello > > I have one doubt. In Drill we can create table using CTAS > Command. > > Like :- CREATE TABLE new_table_name AS <query>; > > > > As Mention in official site:- > > http://drill.apache.org/docs/create-table-as-ctas-command/ > > > > My query is:- Is it possible to update or insert data in CTAS Created > > Table? > > > > > > > > -- > > Thanks & Regards. > > Sanjiv Kumar. > > >
