Netezza -- join columns of the 2 table should be same dataType

 

 

Netezza -- join columns of the 2 table should be same dataType , or there will be data distribution or broadcast

create table test_a (id int) distribute on (id);
create table test_b (id char(10)) distribute on (id);

insert into test_a values (1001);
insert into test_b values ('1001');

 

select count(*),datasliceid from test_a group by datasliceid;
-- 1 42

 

select count(*),datasliceid from test_b group by datasliceid;
-- 2 10

 

select a.id from test_b b , test_a a where a.id = b.id ;

 

QUERY SQL:

explain verbose select a.id from test_b b , test_a a where a.id = b.id

　

　

QUERY VERBOSE PLAN:

Node 1.

[SPU Sequential Scan table "TEST_A" as "A" {(A."ID")}]

-- Estimated Rows = 1, Width = 4, Cost = 0.0 .. 0.0, Conf = 80.0

Restrictions:

(A."ID" NOTNULL)

Projections:

1:A."ID"

[SPU Broadcast]

[HashIt for Join]

Node 2.

[SPU Sequential Scan table "TEST_B" as "B" {(B."ID")}]

-- Estimated Rows = 1, Width = 4, Cost = 0.0 .. 0.0, Conf = 80.0

Restrictions:

(B."ID" NOTNULL)

Projections:

1:INT4((B."ID")::"VARCHAR")

Node 3.

[SPU Hash Join Stream "Node 2" with Temp "Node 1" {(B."ID")}]

-- Estimated Rows = 1, Width = 4, Cost = 0.0 .. 0.0, Conf = 64.0

Restrictions:

(EXPR."INT4((B."ID")::"VARCHAR")" = A."ID")

Projections:

1:A."ID"

[SPU Return]

[Host Return]