PostgreSQL, also known as Postgres, is a free and open-source relational database management system RDBMS emphasizing extensibility and technical standards compliance. It is designed to handle a range of workloads, from single machines to data warehouses or Web services with many concurrent users. It is the default database for macOS Server, and is also available for Linux, FreeBSD, OpenBSD, and Windows.
PostgreSQL features transactions with Atomicity, Consistency, Isolation, Durability ACID properties, automatically updatable views, materialized views, triggers, foreign keys, and stored procedures. PostgreSQL is developed by the PostgreSQL Global Development Group, a diverse group of many companies and individual contributors.
PostgreSQL's developers pronounce PostgreSQL as . It is abbreviated as Postgres because of ubiquitous support for the SQL standard among relational databases. Originally named POSTGRES, the name Post Ingres refers to the project's origins in that RDBMS that originated at University of California, Berkeley. After a review the PostgreSQL Core Team announced in 2007 that the product would continue to use the name PostgreSQL.
PostgreSQL evolved from the Ingres project at the University of California, Berkeley. In 1982, the leader of the Ingres team, Michael Stonebraker, left Berkeley to make a proprietary version of Ingres. He returned to Berkeley in 1985, and began a post-Ingres project to address the problems with contemporary database systems that had become increasingly clear during the early 1980s. He won the Turing Award in 2014 for these and other projects, and techniques pioneered in them.
The new project, POSTGRES, aimed to add the fewest features needed to completely support data types. These features included the ability to define types and to fully describe relationships – something used widely, but maintained entirely by the user. In POSTGRES, the database understood relationships, and could retrieve information in related tables in a natural way using rules. POSTGRES used many of the ideas of Ingres, but not its code.
Starting in 1986, published papers described the basis of the system, and a prototype version was shown at the 1988 ACM ] and an improved query engine. By 1993, the number of users began to overwhelm the project with requests for support and features. After releasing version 4.2 on June 30, 1994 – primarily a cleanup – the project ended. Berkeley released POSTGRES under an MIT License variant, which enabled other developers to use the code for any use. At the time, POSTGRES used an Ingres-influenced POSTQUEL query language interpreter, which could be interactively used with a console application named monitor.
In 1994, Berkeley graduate students Andrew Yu and Jolly Chen replaced the POSTQUEL query language interpreter with one for the SQL query language, creating Postgres95. monitor was also replaced by psql. Yu and Chen announced the first version 0.01 to beta testers on May 5, 1995. Version 1.0 of Postgres95 was announced on September 5, 1995, with a more liberal license that enabled the software to be freely modifiable.
On July 8, 1996, Marc Fournier at Hub.org Networking Services provided the first non-university development server for the open-source development effort. With the participation of Bruce Momjian and Vadim B. Mikheev, work began to stabilize the code inherited from Berkeley.
In 1996, the project was renamed to PostgreSQL to reflect its support for SQL. The online presence at the website PostgreSQL.org began on October 22, 1996. The first PostgreSQL release formed version 6.0 on January 29, 1997. Since then developers and volunteers around the world have maintained the software as The PostgreSQL Global Development Group.
The project continues to make releases available under its free and open-source software PostgreSQL License. Code comes from contributions from proprietary vendors, support companies, and open-source programmers.
PostgreSQL manages concurrency through multiversion concurrency control MVCC, which gives each transaction a "snapshot" of the database, allowing changes to be made without affecting other transactions. This largely eliminates the need for read locks, and ensures the database maintains ACID principles. PostgreSQL offers three levels of transaction isolation: Read Committed, Repeatable Read and Serializable. Because PostgreSQL is immune to dirty reads, requesting a Read Uncommitted transaction isolation level provides read committed instead. PostgreSQL supports full serializability via the serializable snapshot isolation SSI method.
PostgreSQL includes built-in binary replication based on shipping the changes write-ahead logs WAL to replica nodes asynchronously, with the ability to run read-only queries against these replicated nodes. This allows splitting read traffic among multiple nodes efficiently. Earlier replication software that allowed similar read scaling normally relied on adding replication triggers to the master, increasing load.
PostgreSQL includes built-in synchronous replication that ensures that, for each write transaction, the master waits until at least one replica node has written the data to its transaction log. Unlike other database systems, the durability of a transaction whether it is asynchronous or synchronous can be specified per-database, per-user, per-session or even per-transaction. This can be useful for workloads that do not require such guarantees, and may not be wanted for all data as it slows down performance due to the requirement of the confirmation of the transaction reaching the synchronous standby.
Standby servers can be synchronous or asynchronous. Synchronous standby servers can be specified in the configuration which determines which servers are candidates for synchronous replication. The first in the list that is actively streaming will be used as the current synchronous server. When this fails, the system fails over to the next in line.
Synchronous multi-master replication is not included in the PostgreSQL core. Postgres-XC which is based on PostgreSQL provides scalable synchronous multi-master replication. It is licensed under the same license as PostgreSQL. A related project is called Postgres-XL. Postgres-R is yet another fork. Bidirectional replication BDR is an asynchronous multi-master replication system for PostgreSQL.
Tools such as repmgr make managing replication clusters easier.
Several asynchronous trigger-based replication packages are available. These remain useful even after introduction of the expanded core abilities, for situations where binary replication of a full database cluster is inappropriate:
PostgreSQL includes built-in support for regular B-tree and hash table indexes, and four index access methods: generalized search trees GiST, generalized inverted indexes GIN, Space-Partitioned GiST SP-GiST and Block Range Indexes BRIN. In addition, user-defined index methods can be created, although this is quite an involved process. Indexes in PostgreSQL also support the following features:
In PostgreSQL, a schema holds all objects, except for roles and tablespaces. Schemas effectively act like namespaces, allowing objects of the same name to co-exist in the same database. By default, newly created databases have a schema called public, but any further schemas can be added, and the public schema isn't mandatory.
search_path setting determines the order in which PostgreSQL checks schemas for unqualified objects those without a prefixed schema. By default, it is set to
$user refers to the currently connected database user. This default can be set on a database or role level, but as it is a session parameter, it can be freely changed even multiple times during a client session, affecting that session only.
Non-existent schemas listed in search_path are silently skipped during objects lookup.
New objects are created in whichever valid schema one that presently exists appears first in the search_path.Schema is an outline of database.
A wide variety of native data types are supported, including:
In addition, users can create their own data types which can usually be made fully indexable via PostgreSQL's indexing infrastructures – GiST, GIN, SP-GiST. Examples of these include the geographic information system GIS data types from the PostGIS project for PostgreSQL.
There is also a data type called a domain, which is the same as any other data type but with optional constraints defined by the creator of that domain. This means any data entered into a column using the domain will have to conform to whichever constraints were defined as part of the domain.
A data type that represents a range of data can be used which are called range types. These can be discrete ranges e.g. all integer values 1 to 10 or continuous ranges e.g., any time between 10:00 am and 11:00 am. The built-in range types available include ranges of integers, big integers, decimal numbers, time stamps with and without time zone and dates.
Custom range types can be created to make new types of ranges available, such as IP address ranges using the inet type as a base, or float ranges using the float data type as a base. Range types support inclusive and exclusive range boundaries using the [/] and / characters respectively. e.g.,
[4,9 represents all integers starting from and including 4 up to but not including 9. Range types are also compatible with existing operators used to check for overlap, containment, right of etc.
New types of almost all objects inside the database can be created, including:
Tables can be set to inherit their characteristics from a parent table. Data in child tables will appear to exist in the parent tables, unless data is selected from the parent table using the ONLY keyword, i.e.
SELECT * FROM ONLY parent_table;. Adding a column in the parent table will cause that column to appear in the child table.
Inheritance can be used to implement table partitioning, using either triggers or rules to direct inserts to the parent table into the proper child tables.
As of 2010[update], this feature is not fully supported yet – in particular, table constraints are not currently inheritable. All check constraints and not-null constraints on a parent table are automatically inherited by its children. Other types of constraints unique, primary key, and foreign key constraints are not inherited.
Inheritance provides a way to map the features of generalization hierarchies depicted in entity relationship diagrams ERDs directly into the PostgreSQL database.
PostgreSQL can link to other systems to retrieve data via foreign data wrappers FDWs. These can take the form of any data source, such as a file system, another relational database management system RDBMS, or a web service. This means that regular database queries can use these data sources like regular tables, and even join multiple data-sources together.
PostgreSQL has several interfaces available and is also widely supported among programming language libraries. Built-in interfaces include libpq PostgreSQL's official C application interface and ECPG an embedded C system. External interfaces include:
Procedural languages allow developers to extend the database with custom subroutines functions, often called stored procedures. These functions can be used to build database triggers functions invoked on modification of certain data and custom data types and aggregate functions. Procedural languages can also be invoked without defining a function, using a DO command at SQL level.
Languages are divided into two groups: Procedures written in safe languages are sandboxed and can be safely created and used by any user. Procedures written in unsafe languages can only be created by superusers, because they allow bypassing a database's security restrictions, but can also access sources external to the database. Some languages like Perl provide both safe and unsafe versions.
PostgreSQL has built-in support for three procedural languages:
Triggers are events triggered by the action of SQL data manipulation language DML statements. For example, an INSERT statement might activate a trigger that checks if the values of the statement are valid. Most triggers are only activated by either INSERT or UPDATE statements.
Triggers are fully supported and can be attached to tables. Triggers can be per-column and conditional, in that UPDATE triggers can target specific columns of a table, and triggers can be told to execute under a set of conditions as specified in the trigger's WHERE clause. Triggers can be attached to views by using the INSTEAD OF condition. Multiple triggers are fired in alphabetical order. In addition to calling functions written in the native PL/pgSQL, triggers can also invoke functions written in other languages like PL/Python or PL/Perl.
PostgreSQL provides an asynchronous messaging system that is accessed through the NOTIFY, LISTEN and UNLISTEN commands. A session can issue a NOTIFY command, along with the user-specified channel and an optional payload, to mark a particular event occurring. Other sessions are able to detect these events by issuing a LISTEN command, which can listen to a particular channel. This functionality can be used for a wide variety of purposes, such as letting other sessions know when a table has updated or for separate applications to detect when a particular action has been performed. Such a system prevents the need for continuous polling by applications to see if anything has yet changed, and reducing unnecessary overhead. Notifications are fully transactional, in that messages are not sent until the transaction they were sent from is committed. This eliminates the problem of messages being sent for an action being performed which is then rolled back.
Many connectors for PostgreSQL provide support for this notification system including libpq, JDBC, Npgsql, psycopg and node.js so it can be used by external applications.
PostgreSQL can act as an effective, persistent "pub/sub" server or job server by combining LISTEN with FOR UPDATE SKIP LOCKED, a combination which has existed since PostgreSQL version 9.5
Rules allow the "query tree" of an incoming query to be rewritten. "Query Re-Write Rules" are attached to a table/class and "Re-Write" the incoming DML select, insert, update, and/or delete into one or more queries that either replace the original DML statement or execute in addition to it. Query Re-Write occurs after DML statement parsing, but before query planning.
PostgreSQL server is process-based not threaded, and uses one operating system process per database session. Multiple sessions are automatically spread across all available CPUs by the operating system. Starting with PostgreSQL 9.6, many types of queries can also be parallelized across multiple background worker processes, taking advantage of multiple CPUs or cores. Client applications can use threads and create multiple database connections from each thread.
PostgreSQL manages its internal security on a per-role basis. A role is generally regarded to be a user a role that can log in, or a group a role of which other roles are members. Permissions can be granted or revoked on any object down to the column level, and can also allow/prevent the creation of new objects at the database, schema or table levels.
PostgreSQL's SECURITY LABEL feature extension to SQL standards, allows for additional security; with a bundled loadable module that supports label-based mandatory access control MAC based on Security-Enhanced Linux SELinux security policy.
PostgreSQL natively supports a broad number of external authentication mechanisms, including:
The GSSAPI, SSPI, Kerberos, peer, ident and certificate methods can also use a specified "map" file that lists which users matched by that authentication system are allowed to connect as a specific database user.
These methods are specified in the cluster's host-based authentication configuration file pg_hba.conf, which determines what connections are allowed. This allows control over which user can connect to which database, where they can connect from IP address, IP address range, domain socket, which authentication system will be enforced, and whether the connection must use Transport Layer Security TLS.
PostgreSQL claims high, but not complete, conformance with the SQL standard. One exception is the handling of unquoted identifiers like table or column names. In PostgreSQL they are folded – internal – to lower case characters whereas the standard says that unquoted identifiers should be folded to upper case. Thus,
Foo should be equivalent to
foo according to the standard.
Many informal performance studies of PostgreSQL have been done. Performance improvements aimed at improving scalability began heavily with version 8.1. Simple benchmarks between version 8.0 and version 8.4 showed that the latter was more than 10 times faster on read-only workloads and at least 7.5 times faster on both read and write workloads.
The first industry-standard and peer-validated benchmark was completed in June 2007, using the Sun Java System Application Server proprietary version of GlassFish 9.0 Platform Edition, UltraSPARC T1-based Sun Fire server and PostgreSQL 8.2. This result of 778.14 SPECjAppServer2004 JOPS@Standard compares favourably with the 874 JOPS@Standard with Oracle 10 on an Itanium-based HP-UX system.
In August 2007, Sun submitted an improved benchmark score of 813.73 SPECjAppServer2004 JOPS@Standard. With the system under test at a reduced price, the price/performance improved from $84.98/JOPS to $70.57/JOPS.
The default configuration of PostgreSQL uses only a small amount of dedicated memory for performance-critical purposes such as caching database blocks and sorting. This limitation is primarily because older operating systems required kernel changes to allow allocating large blocks of shared memory. PostgreSQL.org provides advice on basic recommended performance practice in a wiki.
In April 2012, Robert Haas of EnterpriseDB demonstrated PostgreSQL 9.2's linear CPU scalability using a server with 64 cores.
Matloob Khushi performed benchmarking between Postgresql 9.0 and MySQL 5.6.15 for their ability to process genomic data. In his performance analysis he found that PostgreSQL extracts overlapping genomic regions eight times faster than MySQL using two datasets of 80,000 each forming random human DNA regions. Insertion and data uploads in PostgreSQL were also better, although general searching ability of both databases was almost equivalent.
PostgreSQL is available for the following operating systems: Linux all recent distributions, 64-bit installers available for macOS OS X version 10.6 and newer – Windows with installers available for 64-bit version; tested on latest versions and back to Windows 2012 R2, while for PostgreSQL version 10 and older a 32-bit installer is available and tested down to 32-bit Windows 2008 R1; compilable by e.g. Visual Studio, version 2013 up up to most recent 2019 version – FreeBSD, OpenBSD, NetBSD, AIX, HP-UX, Solaris, and UnixWare; and not officially tested: DragonFly BSD, BSD/OS, IRIX, OpenIndiana, OpenSolaris, OpenServer, and Tru64 UNIX. Most other Unix-like systems could also work; most modern do support.
PostgreSQL works on any of the following instruction set architectures: x86 and x86-64 on Windows and other operating systems; these are supported on other than Windows: IA-64 Itanium external support for HP-UX, PowerPC, PowerPC 64, S/390, S/390x, SPARC, SPARC 64, ARMv8-A 64-bit and older ARM 32-bit, including older such as ARMv6 in Raspberry Pi, MIPS, MIPSel, and PA-RISC. It was also known to work, but not tested in a while, on Alpha dropped in 9.5, M68k, M32R, NS32k, and VAX. Beyond these, it is possible to build PostgreSQL for an unsupported CPU by disabling spinlocks.
Open source front-ends and tools for administering PostgreSQL include:
A number of companies offer proprietary tools for PostgreSQL. They often consist of a universal core that is adapted for various specific database products. These tools mostly share the administration features with the open source tools but offer improvements in data modeling, importing, exporting or reporting.
Notable organizations and products that use PostgreSQL as the primary database include:
Some notable vendors offer PostgreSQL as software as a service: