EDB Postgres Advanced Server 13 Release Notes¶
With this release of EDB Postgres Advanced Server 13, EnterpriseDB continues to lead as the only worldwide company to deliver innovative and low cost open-source-derived database solutions with commercial quality, ease of use, compatibility, scalability, and performance for small or large-scale enterprises.
EDB Postgres Advanced Server 13 is built on open-source PostgreSQL 13, which introduces myriad enhancements that enable databases to scale up and scale out in more efficient ways. PostgreSQL 13 has significant performance improvements, which includes an improved indexing and lookup system, a level up on query planning when using extended statistics, improved performance for queries that use aggregates or partitioned tables, and providing more ways to monitor activity within a PostgreSQL database. And along with highly requested features like parallelized vacuuming, incremental sorting, PostgreSQL 13 has a multitude of other new features and capabilities.
EDB Postgres Advanced Server 13 adds a number of new features that will delight developers and DBAs alike, including:
Automatic LIST Partitioning
SUBPARTITION numberclauses while creating a HASH partition table
Forward declarations in the package body
CREATE INDEXsyntax contains column name and constant
USING INDEXclause in
EDB Loader handles unique constraint violations
EDB Loader supports any connection parameters with the
Installers and Documentation¶
EDB Postgres Advanced Server v13 is packaged and delivered as interactive installers for Windows; visit the EnterpriseDB website:
RPM Packages are available for Linux from:
Debian/Ubuntu Packages are available for download from:
To request the credentials required to access EDB repositories, visit:
Documentation is provided on the EDB website, visit:
EDB Postgres Advanced Server v13 installers support 64 bit Linux and Windows server platforms. The Advanced Server 13 RPM packages are supported on the following 64-bit Linux platforms:
Red Hat Enterprise Linux (x86_64) 7.x and 8.x
CentOS (x86_64) 7.x and 8.x
OEL (x86_64) 7.x and 8.x
PPC64LE 8 running CentOS/RHEL 7.x
The EDB Postgres Advanced Server 13 native packages are supported on the following 64-bit Linux platforms:
Debian 9.x and 10.x
Ubuntu 18.04 and 20.04
Graphical installers are supported on the following 64-bit Windows platforms:
Windows Server 2016
Additional information about supported platforms is available on the EDB website:
The following components are included in the EDB Postgres Advanced Server v13 release:
Connectors JDBC 188.8.131.52, ODBC 12.02.0000.02, .NET 184.108.40.206, OCL 220.127.116.11
Parallel Clone 1.8
Procedural Language Packs – PL/Perl 5.26, PL/Python 3.7, PL/TCL 8.6
EDB Postgres Advanced Server v13 Features¶
The major highlights of this release are :
Automatic LIST Partitioning is an extension to LIST partitioning that allows a database to automatically create a partition for any new distinct value of the LIST partitioning key. A new partition is created when data is inserted into the LIST partitioned table and the inserted value does not match any of the existing table partitions.
numericvariants in the
Added support for
CREATE INDEXsyntax that contains a column name and number, i.e.
Added support to alter the owner of the directory -
ALTER DIRECTORY <dir_name> OWNER TO <role_name>.
CSV and XML audit logs have been made consistent. It has been observed that there is a difference in terms of number of fields in the CSV and XML formatted audit logs. Now the audit log is consistent across both formats.
Default behaviour for
dbms_outputmade compatible with Redwood. By default, the
offin Oracle; in EPAS it is always
on. In this release, EPAS adds a GUC to control the default behaviour for
You can log the number of processed statements with
edb_log_every_bulk_value. Currently, during bulk execution EPAS did not identify the number of rows processed when
off. This feature allows EPAS to log it both in the audit log file and server log file. This will help you analyze the logs for such cases.
Previously, EDB Loader would abort the whole operation if any record insertion fails due to a unique constraint violation. This is fixed by using speculative insertion to insert rows. This behavior is enforced if
handle_conflicts(a new parameter) is
trueand indexes are present.
SYSDATEnow behaves in a more compatible manner.
SYSDATEoutput changes at every nesting level, so multiple copies of SYSDATE in the same SQL query will return the same value each time.
SYSDATEchanges on successive statements in the same procedure, or within a nested function call.
PARALLEL [n] | NOPARALLELoption for
INDEX. Advanced Server now supports the
PARALLEL [n] | NOPARALLELclause in the
CREATE INDEX, and
ALTER INDEXcommands to enable or disable parallelism on an index or a table.
SUBPARTITION [n]while creating a table. This feature allows you to automatically create
[n]hash partitions at a subpartition level.
EDB Loader supports any connection parameters. You can use the EDB Loader
CONNSTRoptions to specify any connection parameters supported by libpq. This includes SSL connection parameters.
STATS_MODEfunction takes a set of values as an argument and returns the value that occurs with the highest frequency. If multiple values appear with the same frequency, the
STATS_MODEfunction arbitrarily chooses the first value and returns only that one value.
Added support for DBMS_SQL function/procedures (
Added support for function
Added support for function/procedure specification inside package body.
Added support for FM format in
Added support for
Added support for the spell mode in
Allow creating a compound trigger having WHEN clause with NEW/OLD variables and STATEMENT level triggering events. Enhanced Redwood compatible view.
Log matched line of
pg_hba.confon successful client authentication. Enhanced
pg_catcheckto test for and raise an error if a relation’s
relfilenodeis missing from the
Added support for utl_http.end_of_body exception. This feature declares the end_of_body exception into utl_http package and throws the same from read_line, read_text, and read_raw package procedures when no data is left in the response body.
The UNIQUE and PRIMARY KEY constraint clauses now have a CREATE INDEX statement. This new syntax allows users to specify explicit index details like fillfactor, etc. Columns specified in the constraint and the columns specified in the index must be the same.
For information about Advanced Server features that are compatible with Oracle databases, see the following guides:
Database Compatibility for Oracle Developer’s SQL Guide
Database Compatibility for Oracle Developer’s Reference Guide
Database Compatibility for Oracle Developer’s Built-in Package Guide
Database Compatibility for Oracle Developer’s Tools and Utilities Guide
Database Compatibility Table Partitioning Guide
Database Compatibility Stored Procedural Language Guide
Community PostgreSQL 13 Updates¶
EDB Postgres Advanced Server 13 integrates all of the community PostgreSQL 13 features. To review a complete list of changes to the community PostgreSQL project and the contributors names, see the PostgreSQL 13 Release Notes at: