Database Reference
In-Depth Information
he above SQL statement is described below:
◾
bounda r y_va lue
: his value is used to redeine partitions. he type of the value should be
consistent with the data type of the federation key.
◾
SPLIT AT
: At the boundary value, split a federation member to two new federation mem-
bers. he atomic units with the federation key less than the boundary_value are moved to
one of the new federation members and the rest of the atomic units are moved to the other
new federation member.
◾
DROP AT
: Drop the federation members identiied by the boundary_value. When LOW
is used, drop the federation members below the boundary value and extend the range of the
federation members above the boundary_value to include the federation keys of the dropped
federation members. When HIGH is used, drop federation members below the boundary_
value and extend the range of the federation members above the boundary_value to include
the federation keys of the dropped federation members.
SQL Statement for Dropping Federation
: To drop a federation, use the following SQL
statement:
DROP FEDERATION federation_name
SQL Statement for Connecting to Federation
: To connect to a federation, you need to irst
connect to the database served as the federation root. After connect to the federation root, reset
the federation root with the SQL statement below:
USE FEDERATION ROOT WITH RESET
hen, use the following SQL statement to connect to the federation:
USE FEDERATION federation_name (distribution_name
=
value)
WITH FILTERING
=
{ON|OFF}, RESET
he above SQL statement is described below:
◾
value
: It is used to identify the federation member to be connected to.
◾
WITH FILTERING = ON|OFF
: When OFF is set, the entire federation is connected.
When ON is set, the federation member assigned with the value is connected. ON is the
default choice. When ON is set, applications are able to connect to a particular atomic unit
in a federation member automatically.
◾
RESET
: his keyword is required to reset the connection explicitly.
SQL Statement for Creating Federated Table
: After connecting to a federation, the follow-
ing SQL statement can be used to create a federated table:
CREATE TABLE
[schema_name.] table_ame
({
<
column_definition
>
|
<
computed_column_definition
>
}
[
<
table_constraint
>
] [,...n])
FEDERATED ON (distribution_name
=
column_name)
