PostgreSQL supports basic table partitioning. This section describes why and how to implement partitioning as part of your database design.
Partitioning refers to splitting what is logically one large table into smaller physical pieces. Partitioning can provide several benefits:
Query performance can be improved dramatically in certain situations, particularly when most of the heavily accessed rows of the table are in a single partition or a small number of partitions. The partitioning substitutes for leading columns of indexes, reducing index size and making it more likely that the heavily-used parts of the indexes fit in memory.
When queries or updates access a large percentage of a single partition, performance can be improved by taking advantage of sequential scan of that partition instead of using an index and random access reads scattered across the whole table.
Bulk loads and deletes can be accomplished by adding or removing partitions, if that requirement is planned into the partitioning design. ALTER TABLE NO INHERIT and DROP TABLE are both far faster than a bulk operation. These commands also entirely avoid the VACUUM overhead caused by a bulk DELETE.
Seldom-used data can be migrated to cheaper and slower storage media.
The benefits will normally be worthwhile only when a table would otherwise be very large. The exact point at which a table will benefit from partitioning depends on the application, although a rule of thumb is that the size of the table should exceed the physical memory of the database server.
Currently, PostgreSQL supports partitioning via table inheritance. Each partition must be created as a child table of a single parent table. The parent table itself is normally empty; it exists just to represent the entire data set. You should be familiar with inheritance (see Section 5.8) before attempting to set up partitioning.
The following forms of partitioning can be implemented in PostgreSQL:
The table is partitioned into "ranges" defined by a key column or set of columns, with no overlap between the ranges of values assigned to different partitions. For example one might partition by date ranges, or by ranges of identifiers for particular business objects.
The table is partitioned by explicitly listing which key values appear in each partition.
To set up a partitioned table, do the following:
Create the "master" table, from which all of the partitions will inherit.
This table will contain no data. Do not define any check constraints on this table, unless you intend them to be applied equally to all partitions. There is no point in defining any indexes or unique constraints on it, either.
Create several "child" tables that each inherit from the master table. Normally, these tables will not add any columns to the set inherited from the master.
We will refer to the child tables as partitions, though they are in every way normal PostgreSQL tables.
Add table constraints to the partition tables to define the allowed key values in each partition.
Typical examples would be:
CHECK ( x = 1 ) CHECK ( county IN ( 'Oxfordshire', 'Buckinghamshire', 'Warwickshire' )) CHECK ( outletID >= 100 AND outletID < 200 )
Ensure that the constraints guarantee that there is no overlap between the key values permitted in different partitions. A common mistake is to set up range constraints like:
CHECK ( outletID BETWEEN 100 AND 200 ) CHECK ( outletID BETWEEN 200 AND 300 )
This is wrong since it is not clear which partition the key value 200 belongs in.
Note that there is no difference in syntax between range and list partitioning; those terms are descriptive only.
For each partition, create an index on the key column(s), as well as any other indexes you might want. (The key index is not strictly necessary, but in most scenarios it is helpful. If you intend the key values to be unique then you should always create a unique or primary-key constraint for each partition.)
Optionally, define a trigger or rule to redirect data inserted into the master table to the appropriate partition.
Ensure that the constraint_exclusion configuration parameter is not disabled in postgresql.conf. If it is, queries will not be optimized as desired.
For example, suppose we are constructing a database for a large ice cream company. The company measures peak temperatures every day as well as ice cream sales in each region. Conceptually, we want a table like:
CREATE TABLE measurement ( city_id int not null, logdate date not null, peaktemp int, unitsales int );
We know that most queries will access just the last week's, month's or quarter's data, since the main use of this table will be to prepare online reports for management. To reduce the amount of old data that needs to be stored, we decide to only keep the most recent 3 years worth of data. At the beginning of each month we will remove the oldest month's data.
In this situation we can use partitioning to help us meet all of our different requirements for the measurements table. Following the steps outlined above, partitioning can be set up as follows:
The master table is the measurement table, declared exactly as above.
Next we create one partition for each active month:
CREATE TABLE measurement_y2006m02 ( ) INHERITS (measurement); CREATE TABLE measurement_y2006m03 ( ) INHERITS (measurement); ... CREATE TABLE measurement_y2007m11 ( ) INHERITS (measurement); CREATE TABLE measurement_y2007m12 ( ) INHERITS (measurement); CREATE TABLE measurement_y2008m01 ( ) INHERITS (measurement);
Each of the partitions are complete tables in their own right, but they inherit their definitions from the measurement table.
This solves one of our problems: deleting old data. Each month, all we will need to do is perform a DROP TABLE on the oldest child table and create a new child table for the new month's data.
We must provide non-overlapping table constraints. Rather than just creating the partition tables as above, the table creation script should really be:
CREATE TABLE measurement_y2006m02 ( CHECK ( logdate >= DATE '2006-02-01' AND logdate < DATE '2006-03-01' ) ) INHERITS (measurement); CREATE TABLE measurement_y2006m03 ( CHECK ( logdate >= DATE '2006-03-01' AND logdate < DATE '2006-04-01' ) ) INHERITS (measurement); ... CREATE TABLE measurement_y2007m11 ( CHECK ( logdate >= DATE '2007-11-01' AND logdate < DATE '2007-12-01' ) ) INHERITS (measurement); CREATE TABLE measurement_y2007m12 ( CHECK ( logdate >= DATE '2007-12-01' AND logdate < DATE '2008-01-01' ) ) INHERITS (measurement); CREATE TABLE measurement_y2008m01 ( CHECK ( logdate >= DATE '2008-01-01' AND logdate < DATE '2008-02-01' ) ) INHERITS (measurement);
We probably need indexes on the key columns too:
CREATE INDEX measurement_y2006m02_logdate ON measurement_y2006m02 (logdate); CREATE INDEX measurement_y2006m03_logdate ON measurement_y2006m03 (logdate); ... CREATE INDEX measurement_y2007m11_logdate ON measurement_y2007m11 (logdate); CREATE INDEX measurement_y2007m12_logdate ON measurement_y2007m12 (logdate); CREATE INDEX measurement_y2008m01_logdate ON measurement_y2008m01 (logdate);
We choose not to add further indexes at this time.
We want our application to be able to say INSERT INTO measurement ... and have the data be redirected into the appropriate partition table. We can arrange that by attaching a suitable trigger function to the master table. If data will be added only to the latest partition, we can use a very simple trigger function:
CREATE OR REPLACE FUNCTION measurement_insert_trigger() RETURNS TRIGGER AS $$ BEGIN INSERT INTO measurement_y2008m01 VALUES (NEW.*); RETURN NULL; END; $$ LANGUAGE plpgsql;
After creating the function, we create a trigger which calls the trigger function:
CREATE TRIGGER insert_measurement_trigger BEFORE INSERT ON measurement FOR EACH ROW EXECUTE PROCEDURE measurement_insert_trigger();
We must redefine the trigger function each month so that it always points to the current partition. The trigger definition does not need to be updated, however.
We might want to insert data and have the server automatically locate the partition into which the row should be added. We could do this with a more complex trigger function, for example:
CREATE OR REPLACE FUNCTION measurement_insert_trigger() RETURNS TRIGGER AS $$ BEGIN IF ( NEW.logdate >= DATE '2006-02-01' AND NEW.logdate < DATE '2006-03-01' ) THEN INSERT INTO measurement_y2006m02 VALUES (NEW.*); ELSIF ( NEW.logdate >= DATE '2006-03-01' AND NEW.logdate < DATE '2006-04-01' ) THEN INSERT INTO measurement_y2006m03 VALUES (NEW.*); ... ELSIF ( NEW.logdate >= DATE '2008-01-01' AND NEW.logdate < DATE '2008-02-01' ) THEN INSERT INTO measurement_y2008m01 VALUES (NEW.*); ELSE RAISE EXCEPTION 'Date out of range. Fix the measurement_insert_trigger() function!'; END IF; RETURN NULL; END; $$ LANGUAGE plpgsql;
The trigger definition is the same as before. Note that each IF test must exactly match the CHECK constraint for its partition.
While this function is more complex than the single-month case, it doesn't need to be updated as often, since branches can be added in advance of being needed.
Note: In practice it might be best to check the newest partition first, if most inserts go into that partition. For simplicity we have shown the trigger's tests in the same order as in other parts of this example.
As we can see, a complex partitioning scheme could require a substantial amount of DDL. In the above example we would be creating a new partition each month, so it might be wise to write a script that generates the required DDL automatically.
Normally the set of partitions established when initially defining the table are not intended to remain static. It is common to want to remove old partitions of data and periodically add new partitions for new data. One of the most important advantages of partitioning is precisely that it allows this otherwise painful task to be executed nearly instantaneously by manipulating the partition structure, rather than physically moving large amounts of data around.
The simplest option for removing old data is simply to drop the partition that is no longer necessary:
DROP TABLE measurement_y2006m02;
This can very quickly delete millions of records because it doesn't have to individually delete every record.
Another option that is often preferable is to remove the partition from the partitioned table but retain access to it as a table in its own right:
ALTER TABLE measurement_y2006m02 NO INHERIT measurement;
This allows further operations to be performed on the data before it is dropped. For example, this is often a useful time to back up the data using COPY, pg_dump, or similar tools. It might also be a useful time to aggregate data into smaller formats, perform other data manipulations, or run reports.
Similarly we can add a new partition to handle new data. We can create an empty partition in the partitioned table just as the original partitions were created above:
CREATE TABLE measurement_y2008m02 ( CHECK ( logdate >= DATE '2008-02-01' AND logdate < DATE '2008-03-01' ) ) INHERITS (measurement);
As an alternative, it is sometimes more convenient to create the new table outside the partition structure, and make it a proper partition later. This allows the data to be loaded, checked, and transformed prior to it appearing in the partitioned table:
CREATE TABLE measurement_y2008m02 (LIKE measurement INCLUDING DEFAULTS INCLUDING CONSTRAINTS); ALTER TABLE measurement_y2008m02 ADD CONSTRAINT y2008m02 CHECK ( logdate >= DATE '2008-02-01' AND logdate < DATE '2008-03-01' ); \copy measurement_y2008m02 from 'measurement_y2008m02' -- possibly some other data preparation work ALTER TABLE measurement_y2008m02 INHERIT measurement;
Constraint exclusion is a query optimization technique that improves performance for partitioned tables defined in the fashion described above. As an example:
SET constraint_exclusion = on; SELECT count(*) FROM measurement WHERE logdate >= DATE '2008-01-01';
Without constraint exclusion, the above query would scan each of the partitions of the measurement table. With constraint exclusion enabled, the planner will examine the constraints of each partition and try to prove that the partition need not be scanned because it could not contain any rows meeting the query's WHERE clause. When the planner can prove this, it excludes the partition from the query plan.
You can use the EXPLAIN command to show the difference between a plan with constraint_exclusion on and a plan with it off. A typical unoptimized plan for this type of table setup is:
SET constraint_exclusion = off; EXPLAIN SELECT count(*) FROM measurement WHERE logdate >= DATE '2008-01-01'; QUERY PLAN ----------------------------------------------------------------------------------------------- Aggregate (cost=158.66..158.68 rows=1 width=0) -> Append (cost=0.00..151.88 rows=2715 width=0) -> Seq Scan on measurement (cost=0.00..30.38 rows=543 width=0) Filter: (logdate >= '2008-01-01'::date) -> Seq Scan on measurement_y2006m02 measurement (cost=0.00..30.38 rows=543 width=0) Filter: (logdate >= '2008-01-01'::date) -> Seq Scan on measurement_y2006m03 measurement (cost=0.00..30.38 rows=543 width=0) Filter: (logdate >= '2008-01-01'::date) ... -> Seq Scan on measurement_y2007m12 measurement (cost=0.00..30.38 rows=543 width=0) Filter: (logdate >= '2008-01-01'::date) -> Seq Scan on measurement_y2008m01 measurement (cost=0.00..30.38 rows=543 width=0) Filter: (logdate >= '2008-01-01'::date)
Some or all of the partitions might use index scans instead of full-table sequential scans, but the point here is that there is no need to scan the older partitions at all to answer this query. When we enable constraint exclusion, we get a significantly cheaper plan that will deliver the same answer:
SET constraint_exclusion = on; EXPLAIN SELECT count(*) FROM measurement WHERE logdate >= DATE '2008-01-01'; QUERY PLAN ----------------------------------------------------------------------------------------------- Aggregate (cost=63.47..63.48 rows=1 width=0) -> Append (cost=0.00..60.75 rows=1086 width=0) -> Seq Scan on measurement (cost=0.00..30.38 rows=543 width=0) Filter: (logdate >= '2008-01-01'::date) -> Seq Scan on measurement_y2008m01 measurement (cost=0.00..30.38 rows=543 width=0) Filter: (logdate >= '2008-01-01'::date)
Note that constraint exclusion is driven only by CHECK constraints, not by the presence of indexes. Therefore it isn't necessary to define indexes on the key columns. Whether an index needs to be created for a given partition depends on whether you expect that queries that scan the partition will generally scan a large part of the partition or just a small part. An index will be helpful in the latter case but not the former.
The default (and recommended) setting of constraint_exclusion is actually neither on nor off, but an intermediate setting called partition, which causes the technique to be applied only to queries that are likely to be working on partitioned tables. The on setting causes the planner to examine CHECK constraints in all queries, even simple ones that are unlikely to benefit.
A different approach to redirecting inserts into the appropriate partition table is to set up rules, instead of a trigger, on the master table. For example:
CREATE RULE measurement_insert_y2006m02 AS ON INSERT TO measurement WHERE ( logdate >= DATE '2006-02-01' AND logdate < DATE '2006-03-01' ) DO INSTEAD INSERT INTO measurement_y2006m02 VALUES (NEW.*); ... CREATE RULE measurement_insert_y2008m01 AS ON INSERT TO measurement WHERE ( logdate >= DATE '2008-01-01' AND logdate < DATE '2008-02-01' ) DO INSTEAD INSERT INTO measurement_y2008m01 VALUES (NEW.*);
A rule has significantly more overhead than a trigger, but the overhead is paid once per query rather than once per row, so this method might be advantageous for bulk-insert situations. In most cases, however, the trigger method will offer better performance.
Be aware that COPY ignores rules. If you want to use COPY to insert data, you'll need to copy into the correct partition table rather than into the master. COPY does fire triggers, so you can use it normally if you use the trigger approach.
Another disadvantage of the rule approach is that there is no simple way to force an error if the set of rules doesn't cover the insertion date; the data will silently go into the master table instead.
Partitioning can also be arranged using a UNION ALL view, instead of table inheritance. For example,
CREATE VIEW measurement AS SELECT * FROM measurement_y2006m02 UNION ALL SELECT * FROM measurement_y2006m03 ... UNION ALL SELECT * FROM measurement_y2007m11 UNION ALL SELECT * FROM measurement_y2007m12 UNION ALL SELECT * FROM measurement_y2008m01;
However, the need to recreate the view adds an extra step to adding and dropping individual partitions of the data set. In practice this method has little to recommend it compared to using inheritance.
The following caveats apply to partitioned tables:
There is no automatic way to verify that all of the CHECK constraints are mutually exclusive. It is safer to create code that generates partitions and creates and/or modifies associated objects than to write each by hand.
The schemes shown here assume that the partition key column(s) of a row never change, or at least do not change enough to require it to move to another partition. An UPDATE that attempts to do that will fail because of the CHECK constraints. If you need to handle such cases, you can put suitable update triggers on the partition tables, but it makes management of the structure much more complicated.
If you are using manual VACUUM or ANALYZE commands, don't forget that you need to run them on each partition individually. A command like:
will only process the master table.
The following caveats apply to constraint exclusion:
Constraint exclusion only works when the query's
WHERE clause contains constants. A
parameterized query will not be optimized, since the
planner cannot know which partitions the parameter value
might select at run time. For the same reason, "stable" functions such as
CURRENT_DATE must be avoided.
Keep the partitioning constraints simple, else the planner may not be able to prove that partitions don't need to be visited. Use simple equality conditions for list partitioning, or simple range tests for range partitioning, as illustrated in the preceding examples. A good rule of thumb is that partitioning constraints should contain only comparisons of the partitioning column(s) to constants using B-tree-indexable operators.
All constraints on all partitions of the master table are examined during constraint exclusion, so large numbers of partitions are likely to increase query planning time considerably. Partitioning using these techniques will work well with up to perhaps a hundred partitions; don't try to use many thousands of partitions.
Here's a simple way to automate partition creation.
CREATE OR REPLACE FUNCTION logs_insert_func()
RETURNS TRIGGER AS $$
-- The table we'll inherit from
ourMasterTable := 'logs';
-- Get the partition table names ~ master_year_month
SELECT ourMasterTable|| '_' || EXTRACT(ISOYEAR FROM NEW.log_time) || '_' || EXTRACT(MONTH FROM NEW.log_time) into ourTable;
-- Create our insert statement
ourInsertSTMT := 'INSERT INTO '|| ourTable || ' (status,log_time,svc_time,ip_addr,query) VALUES (';
ourInsertSTMT := ourInsertSTMT|| NEW.status ||',';
ourInsertSTMT := ourInsertSTMT|| quote_nullable(NEW.log_time) ||',';
ourInsertSTMT := ourInsertSTMT|| NEW.svc_time ||',';
ourInsertSTMT := ourInsertSTMT|| quote_nullable(NEW.ip_addr) ||',';
ourInsertSTMT := ourInsertSTMT|| quote_nullable(NEW.query);
ourInsertSTMT := ourInsertSTMT || ')';
--Try execute it
-- Phew! We didn't Except
WHEN OTHERS THEN
-- Insert failed. Let's check whether the table exists
SELECT count(*) into ourTableExists
FROM pg_catalog.pg_class c
WHERE c.relname = ourTable;
-- If it doesn't exist, try create it
IF ourTableExists = 0 THEN
-- First of this month and next month
SELECT date_trunc('month', NEW.log_time) into ourFirstOfMonth;
SELECT (ourFirstOfMonth + interval '1 month')::date into ourFirstOfNextMonth;
-- Create partition with range
ourCreateSTMT := 'CREATE TABLE '|| ourTable || '(';
ourCreateSTMT := ourCreateSTMT ||' CHECK ( log_time >=' || quote_nullable(ourFirstOfMonth) ;
ourCreateSTMT := ourCreateSTMT ||' AND log_time < DATE '|| quote_nullable(ourFirstOfNextMonth) || ')';
ourCreateSTMT := ourCreateSTMT ||') INHERITS (' || ourMasterTable || ')';
RAISE NOTICE 'Attempting to create a new table with STMT %',ourCreateSTMT;
-- Retry to insert row
IF NOT found THEN
RAISE NOTICE 'Error inserting into created partition % for %',ourTable,ourInsertSTMT;
RAISE NOTICE 'Error inserting into existing partition % for %',ourTable,ourInsertSTMT;
The master table:
CREATE TABLE logs
id bigserial NOT NULL,
log_time timestamp without time zone,
query character varying(2048),
CONSTRAINT logs_id PRIMARY KEY (id)
CREATE TRIGGER logs_insert_trigger
FOR EACH ROW
EXECUTE PROCEDURE logs_insert_func();
=# insert into logs (status,log_time,svc_time,ip_addr,query) (select status,log_time,svc_time,ip_addr,query from whois_logs);
NOTICE: Attempting to create a new table with STMT CREATE TABLE logs_2009_1( CHECK ( log_time >='2010-01-01' AND log_time < DATE '2010-02-01')) INHERITS (logs)
NOTICE: Attempting to create a new table with STMT CREATE TABLE logs_2009_5( CHECK ( log_time >='2009-05-01' AND log_time < DATE '2009-06-01')) INHERITS (logs)
NOTICE: Attempting to create a new table with STMT CREATE TABLE logs_2006_9( CHECK ( log_time >='2006-09-01' AND log_time < DATE '2006-10-01')) INHERITS (logs)