1 SQL Server에서 ODBC연결하여 Hive에서 데이터 가져올때

2 파티션 컬럼 읽을 때

3 자원 쓰기

drop table if exists #temp1
select *
into #temp1
from openquery(hive, '
select
    col1, col2, col3
from gamelog
where 1=1
and a.date_key between 20190619 and 20190821
'
)

exec('drop table if exists etl_fcc1d2fd1da44550bbde0661f25b631a;') at hive
 
exec('
create table etl_fcc1d2fd1da44550bbde0661f25b631a
as
select
    col1, col2, col3
from gamelog
where 1=1
and a.date_key between 20190619 and 20190821
') at hive
go
 
drop table if exists #1
select *
into #1
from openquery(hive, '
select
    col1, col2, col3
from etl_fcc1d2fd1da44550bbde0661f25b631a')
--4:30

hive> set hive.execution.engine=mr;
hive> select log_no
    > from gamelog
    > where 1=1
    > and date_key=20181228
    > group by log_no;

hive> select log_no, count(get_json_object(log_string, '$.1') cnt
    > from gamelog
    > where 1=1
    > and date_key=20181228
    > group by log_no;

set hive.mapred.reduce.tasks.speculative.execution =true;

set mapreduce.input.fileinputformat.split.maxsize=1048576;