PostgreSQL devises a query plan for each query it is given. Choosing the right plan to match the query structure and the properties of the data is absolutely critical for good performance, so the system includes a complex planner that tries to select good plans. You can use the EXPLAIN command to see what query plan the planner creates for any query. Plan-reading is an art that deserves an extensive tutorial, which this is not; but here is some basic information.
The structure of a query plan is a tree of plan nodes. Nodes at the bottom level are table scan nodes: they return raw rows from a table. There are different types of scan nodes for different table access methods: sequential scans, index scans, and bitmap index scans. If the query requires joining, aggregation, sorting, or other operations on the raw rows, then there will be additional nodes "atop" the scan nodes to perform these operations. Again, there is usually more than one possible way to do these operations, so different node types can appear here too. The output of EXPLAIN has one line for each node in the plan tree, showing the basic node type plus the cost estimates that the planner made for the execution of that plan node. The first line (topmost node) has the estimated total execution cost for the plan; it is this number that the planner seeks to minimize.
Here is a trivial example, just to show what the output looks like. [1]
EXPLAIN SELECT * FROM tenk1;
QUERY PLAN
-------------------------------------------------------------
Seq Scan on tenk1 (cost=0.00..458.00 rows=10000 width=244)
The numbers that are quoted by EXPLAIN are:
Estimated start-up cost (Time expended before output scan can start, e.g., time to do the sorting in a sort node.)
Estimated total cost (If all rows were to be retrieved, which they may not be: for example, a query with a LIMIT clause will stop short of paying the total cost of the Limit plan node's input node.)
Estimated number of rows output by this plan node (Again, only if executed to completion.)
Estimated average width (in bytes) of rows output by this plan node
The costs are measured in arbitrary units determined by the planner's cost parameters (see Seção 17.6.2). Traditional practice is to measure the costs in units of disk page fetches; that is, seq_page_cost is conventionally set to 1.0 and the other cost parameters are set relative to that. The examples in this section are run with the default cost parameters.
It's important to note that the cost of an upper-level node includes the cost of all its child nodes. It's also important to realize that the cost only reflects things that the planner cares about. In particular, the cost does not consider the time spent transmitting result rows to the client, which could be an important factor in the true elapsed time; but the planner ignores it because it cannot change it by altering the plan. (Every correct plan will output the same row set, we trust.)
Rows output is a little tricky because it is not the number of rows processed or scanned by the plan node. It is usually less, reflecting the estimated selectivity of any WHERE-clause conditions that are being applied at the node. Ideally the top-level rows estimate will approximate the number of rows actually returned, updated, or deleted by the query.
Returning to our example:
EXPLAIN SELECT * FROM tenk1;
QUERY PLAN
-------------------------------------------------------------
Seq Scan on tenk1 (cost=0.00..458.00 rows=10000 width=244)
This is about as straightforward as it gets. If you do
SELECT relpages, reltuples FROM pg_class WHERE relname = 'tenk1';
you will find out that tenk1 has 358 disk pages and 10000 rows. So the cost is estimated at 358 page reads, costing seq_page_cost apiece (1.0 by default), plus 10000 * cpu_tuple_cost which is 0.01 by default.
Now let's modify the query to add a WHERE condition:
EXPLAIN SELECT * FROM tenk1 WHERE unique1 < 7000;
QUERY PLAN
------------------------------------------------------------
Seq Scan on tenk1 (cost=0.00..483.00 rows=7033 width=244)
Filter: (unique1 < 7000)
Notice that the EXPLAIN output shows the WHERE clause being applied as a "filter" condition; this means that the plan node checks the condition for each row it scans, and outputs only the ones that pass the condition. The estimate of output rows has gone down because of the WHERE clause. However, the scan will still have to visit all 10000 rows, so the cost hasn't decreased; in fact it has gone up a bit to reflect the extra CPU time spent checking the WHERE condition.
The actual number of rows this query would select is 7000, but the rows estimate is only approximate. If you try to duplicate this experiment, you will probably get a slightly different estimate; moreover, it will change after each ANALYZE command, because the statistics produced by ANALYZE are taken from a randomized sample of the table.
Now, let's make the condition more restrictive:
EXPLAIN SELECT * FROM tenk1 WHERE unique1 < 100;
QUERY PLAN
------------------------------------------------------------------------------
Bitmap Heap Scan on tenk1 (cost=2.37..232.35 rows=106 width=244)
Recheck Cond: (unique1 < 100)
-> Bitmap Index Scan on tenk1_unique1 (cost=0.00..2.37 rows=106 width=0)
Index Cond: (unique1 < 100)
Here the planner has decided to use a two-step plan: the bottom plan node visits an index to find the locations of rows matching the index condition, and then the upper plan node actually fetches those rows from the table itself. Fetching the rows separately is much more expensive than sequentially reading them, but because not all the pages of the table have to be visited, this is still cheaper than a sequential scan. (The reason for using two levels of plan is that the upper plan node sorts the row locations identified by the index into physical order before reading them, so as to minimize the costs of the separate fetches. The "bitmap" mentioned in the node names is the mechanism that does the sorting.)
If the WHERE condition is selective enough, the planner may switch to a "simple" index scan plan:
EXPLAIN SELECT * FROM tenk1 WHERE unique1 < 3;
QUERY PLAN
------------------------------------------------------------------------------
Index Scan using tenk1_unique1 on tenk1 (cost=0.00..10.00 rows=2 width=244)
Index Cond: (unique1 < 3)
In this case the table rows are fetched in index order, which makes them even more expensive to read, but there are so few that the extra cost of sorting the row locations is not worth it. You'll most often see this plan type for queries that fetch just a single row, and for queries that request an ORDER BY condition that matches the index order.
Add another condition to the WHERE clause:
EXPLAIN SELECT * FROM tenk1 WHERE unique1 < 3 AND stringu1 = 'xxx';
QUERY PLAN
------------------------------------------------------------------------------
Index Scan using tenk1_unique1 on tenk1 (cost=0.00..10.01 rows=1 width=244)
Index Cond: (unique1 < 3)
Filter: (stringu1 = 'xxx'::name)
The added condition stringu1 = 'xxx' reduces the output-rows estimate, but not the cost because we still have to visit the same set of rows. Notice that the stringu1 clause cannot be applied as an index condition (since this index is only on the unique1 column). Instead it is applied as a filter on the rows retrieved by the index. Thus the cost has actually gone up a little bit to reflect this extra checking.
If there are indexes on several columns used in WHERE, the planner might choose to use an AND or OR combination of the indexes:
EXPLAIN SELECT * FROM tenk1 WHERE unique1 < 100 AND unique2 > 9000;
QUERY PLAN
-------------------------------------------------------------------------------------
Bitmap Heap Scan on tenk1 (cost=11.27..49.11 rows=11 width=244)
Recheck Cond: ((unique1 < 100) AND (unique2 > 9000))
-> BitmapAnd (cost=11.27..11.27 rows=11 width=0)
-> Bitmap Index Scan on tenk1_unique1 (cost=0.00..2.37 rows=106 width=0)
Index Cond: (unique1 < 100)
-> Bitmap Index Scan on tenk1_unique2 (cost=0.00..8.65 rows=1042 width=0)
Index Cond: (unique2 > 9000)
But this requires visiting both indexes, so it's not necessarily a win compared to using just one index and treating the other condition as a filter. If you vary the ranges involved you'll see the plan change accordingly.
Let's try joining two tables, using the columns we have been discussing:
EXPLAIN SELECT * FROM tenk1 t1, tenk2 t2 WHERE t1.unique1 < 100 AND t1.unique2 = t2.unique2;
QUERY PLAN
--------------------------------------------------------------------------------------
Nested Loop (cost=2.37..553.11 rows=106 width=488)
-> Bitmap Heap Scan on tenk1 t1 (cost=2.37..232.35 rows=106 width=244)
Recheck Cond: (unique1 < 100)
-> Bitmap Index Scan on tenk1_unique1 (cost=0.00..2.37 rows=106 width=0)
Index Cond: (unique1 < 100)
-> Index Scan using tenk2_unique2 on tenk2 t2 (cost=0.00..3.01 rows=1 width=244)
Index Cond: ("outer".unique2 = t2.unique2)
In this nested-loop join, the outer scan is the same bitmap index scan we saw earlier, and so its cost and row count are the same because we are applying the WHERE clause unique1 < 100 at that node. The t1.unique2 = t2.unique2 clause is not relevant yet, so it doesn't affect row count of the outer scan. For the inner scan, the unique2 value of the current outer-scan row is plugged into the inner index scan to produce an index condition like t2.unique2 = constante. So we get the same inner-scan plan and costs that we'd get from, say, EXPLAIN SELECT * FROM tenk2 WHERE unique2 = 42. The costs of the loop node are then set on the basis of the cost of the outer scan, plus one repetition of the inner scan for each outer row (106 * 3.01, here), plus a little CPU time for join processing.
In this example the join's output row count is the same as the product of the two scans' row counts, but that's not true in general, because in general you can have WHERE clauses that mention both tables and so can only be applied at the join point, not to either input scan. For example, if we added WHERE ... AND t1.hundred < t2.hundred, that would decrease the output row count of the join node, but not change either input scan.
One way to look at variant plans is to force the planner to disregard whatever strategy it thought was the winner, using the enable/disable flags described in Seção 17.6.1. (This is a crude tool, but useful. See also Seção 13.3.)
SET enable_nestloop = off;
EXPLAIN SELECT * FROM tenk1 t1, tenk2 t2 WHERE t1.unique1 < 100 AND t1.unique2 = t2.unique2;
QUERY PLAN
------------------------------------------------------------------------------------------
Hash Join (cost=232.61..741.67 rows=106 width=488)
Hash Cond: ("outer".unique2 = "inner".unique2)
-> Seq Scan on tenk2 t2 (cost=0.00..458.00 rows=10000 width=244)
-> Hash (cost=232.35..232.35 rows=106 width=244)
-> Bitmap Heap Scan on tenk1 t1 (cost=2.37..232.35 rows=106 width=244)
Recheck Cond: (unique1 < 100)
-> Bitmap Index Scan on tenk1_unique1 (cost=0.00..2.37 rows=106 width=0)
Index Cond: (unique1 < 100)
This plan proposes to extract the 100 interesting rows of tenk1 using that same old index scan, stash them into an in-memory hash table, and then do a sequential scan of tenk2, probing into the hash table for possible matches of t1.unique2 = t2.unique2 at each tenk2 row. The cost to read tenk1 and set up the hash table is entirely start-up cost for the hash join, since we won't get any rows out until we can start reading tenk2. The total time estimate for the join also includes a hefty charge for the CPU time to probe the hash table 10000 times. Note, however, that we are not charging 10000 times 232.35; the hash table setup is only done once in this plan type.
It is possible to check on the accuracy of the planner's estimated costs by using EXPLAIN ANALYZE. This command actually executes the query, and then displays the true run time accumulated within each plan node along with the same estimated costs that a plain EXPLAIN shows. For example, we might get a result like this:
EXPLAIN ANALYZE SELECT * FROM tenk1 t1, tenk2 t2 WHERE t1.unique1 < 100 AND t1.unique2 = t2.unique2;
QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------
Nested Loop (cost=2.37..553.11 rows=106 width=488) (actual time=1.392..12.700 rows=100 loops=1)
-> Bitmap Heap Scan on tenk1 t1 (cost=2.37..232.35 rows=106 width=244) (actual time=0.878..2.367 rows=100 loops=1)
Recheck Cond: (unique1 < 100)
-> Bitmap Index Scan on tenk1_unique1 (cost=0.00..2.37 rows=106 width=0) (actual time=0.546..0.546 rows=100 loops=1)
Index Cond: (unique1 < 100)
-> Index Scan using tenk2_unique2 on tenk2 t2 (cost=0.00..3.01 rows=1 width=244) (actual time=0.067..0.078 rows=1 loops=100)
Index Cond: ("outer".unique2 = t2.unique2)
Total runtime: 14.452 ms
Note that the "actual time" values are in milliseconds of real time, whereas the "cost" estimates are expressed in arbitrary units; so they are unlikely to match up. The thing to pay attention to is whether the ratios of actual time and estimated costs are consistent.
In some query plans, it is possible for a subplan node to be executed more than once. For example, the inner index scan is executed once per outer row in the above nested-loop plan. In such cases, the "loops" value reports the total number of executions of the node, and the actual time and rows values shown are averages per-execution. This is done to make the numbers comparable with the way that the cost estimates are shown. Multiply by the "loops" value to get the total time actually spent in the node.
The Total runtime shown by EXPLAIN ANALYZE includes executor start-up and shut-down time, as well as time spent processing the result rows. It does not include parsing, rewriting, or planning time. For a SELECT query, the total run time will normally be just a little larger than the total time reported for the top-level plan node. For INSERT, UPDATE, and DELETE commands, the total run time may be considerably larger, because it includes the time spent processing the result rows. In these commands, the time for the top plan node essentially is the time spent computing the new rows and/or locating the old ones, but it doesn't include the time spent applying the changes. Time spent firing triggers, if any, is also outside the top plan node, and is shown separately for each trigger.
It is worth noting that EXPLAIN results should not be extrapolated to situations other than the one you are actually testing; for example, results on a toy-sized table can't be assumed to apply to large tables. The planner's cost estimates are not linear and so it may well choose a different plan for a larger or smaller table. An extreme example is that on a table that only occupies one disk page, you'll nearly always get a sequential scan plan whether indexes are available or not. The planner realizes that it's going to take one disk page read to process the table in any case, so there's no value in expending additional page reads to look at an index.
| [1] |
Examples in this section are drawn from the regression test database after doing a VACUUM ANALYZE, using 8.2 development sources. You should be able to get similar results if you try the examples yourself, but your estimated costs and row counts will probably vary slightly because ANALYZE's statistics are random samples rather than being exact. |
