Babelfish for Aurora PostgreSQL procedure reference
Overview
You can use the following procedure for Amazon RDS DB instances running Babelfish for Aurora PostgreSQL for a better query performance:
sp_babelfish_volatility
PostgreSQL function volatility helps the optimizer for a better query execution which when used in parts of certain clauses has a significant impact on query performance.
Syntax
sp_babelfish_volatility ‘function_name’, ‘volatility’
Arguments
- function_name (optional)
-
You can either specify the value of this argument with a two-part name as
schema_name.function_name
or only thefunction_name
. If you specify only thefunction_name
, the schema name is the default schema for the current user. - volatility (optional)
-
The valid PostgreSQL values of volatility are
stable
,volatile
, orimmutable
. For more information, see https://www.postgresql.org/docs/current/xfunc-volatility.html
Note
When sp_babelfish_volatility
is called with function_name
which has multiple definitions, it will throw an error.
Result set
If the parameters are not mentioned then the result set is displayed under the following columns: schemaname
, functionname
, volatility
.
Usage notes
PostgreSQL function volatility helps the optimizer for a better query execution which when used in parts of certain clauses has a significant impact on query performance.
Examples
The following examples shows how to create simple functions and later explains how to use sp_babelfish_volatility
on these functions
using different methods.
1>
create function f1() returns int as begin return 0 end2>
go
1>
create schema test_schema2>
go
1>
create function test_schema.f1() returns int as begin return 0 end2>
go
The following example displays volatility of the functions:
1>
exec sp_babelfish_volatility2>
goschemaname functionname volatility ----------- ------------ ---------- dbo f1 volatile test_schema f1 volatile
The following example shows how to change the volatility of the functions:
1>
exec sp_babelfish_volatility 'f1','stable'2>
go1>
exec sp_babelfish_volatility 'test_schema.f1','immutable'2>
go
When you specify only the function_name, it displays the schema name, function name and volatility of that function. The following example displays volatility of functions after changing the values:
1>
exec sp_babelfish_volatility 'test_schema.f1'2>
goschemaname functionname volatility ----------- ------------ ---------- test_schema f1 immutable
1>
exec sp_babelfish_volatility 'f1'2>
goschemaname functionname volatility ----------- ------------ ---------- dbo f1 stable
When you don't specify any argument, it displays a list of functions (schema name, function name, volatility of the functions) present in the current database:
1>
exec sp_babelfish_volatility2>
goschemaname functionname volatility ----------- ------------ ---------- dbo f1 stable test_schema f1 immutable
sp_execute_postgresql
You can execute PostgreSQL statements from T-SQL endpoint. This simplifies your applications as you don’t need to exit T-SQL port to execute these statements.
Syntax
sp_execute_postgresql [ @stmt = ] statement
Arguments
- [ @stmt ] statement
-
The argument is of datatype varchar. This argument accept PG dialect statements.
Note
You can only pass one PG dialect statement as an argument otherwise it will raise the following error.
1>
exec sp_execute_postgresql 'create extension pg_stat_statements; drop extension pg_stat_statements'2>
go
Msg 33557097, Level 16, State 1, Server BABELFISH, Line 1 expected 1 statement but got 2 statements after parsing
Usage notes
CREATE EXTENSION
Creates and loads a new extension into the current database.
1>
EXEC sp_execute_postgresql 'create extension [ IF NOT EXISTS ]<extension name>
[ WITH ] [SCHEMAschema_name
] [VERSIONversion
]';2>
go
The following example shows how to create an extension:
1>
EXEC sp_execute_postgresql 'create extension pg_stat_statements with schema sys version "1.10"';2>
go
Use the following command to access extension objects:
1>
select * from pg_stat_statements;2>
go
Note
If schema name is not provided explicitly during extension creation, by default the extensions are installed in the public schema. You must provide the schema qualifier to access the extension objects as mentioned below:
1>
select * from [public].pg_stat_statements;2>
go
Supported extensions
The following extensions available with Aurora PostgreSQL works with Babelfish.
pg_stat_statements
tds_fdw
fuzzystrmatch
Limitations
Users need to have sysadmin role on T-SQL and rds_superuser on postgres to install the extenstions.
Extensions cannot be installed in user created schemas and also in dbo and guest schemas for master, tempdb and msdb database.
CASCADE option is not supported.
ALTER EXTENSION
You can upgrade to a new extension version using ALTER extension.
1>
EXEC sp_execute_postgresql 'alter extension <extension name> UPDATE TO <new_version>';2>
go
Limitations
You can upgrade the version of your extension only using the ALTER Extension statement. Other operations aren't supported.
DROP EXTENSION
Drops the specified extension. You can also use if exists
or restrict
options to drop the extension.
1>
EXEC sp_execute_postgresql 'drop extension <extension name>';2>
go
Limitations
CASCADE option is not supported.