Using PostgreSQL (8.4), I'm creating a view that summarizes various results from a few tables (e.g creating columns a, b, c in the view), and then I need to combine some of these results together in the same query (e.g. a+b, a-b, (a+b)/c, ...), so as to produce the final results. What I'm noticing is that the intermediate results are fully computed each time they are used, even if it's done within the same query.
Is there a way to optimize this so as to avoid those same results to be computed every time?
Here is a simplified example that reproduces the problem.
CREATE TABLE test1 (
id SERIAL PRIMARY KEY,
log_timestamp TIMESTAMP NOT NULL
);
CREATE TABLE test2 (
test1_id INTEGER NOT NULL REFERENCES test1(id),
category VARCHAR(10) NOT NULL,
col1 INTEGER,
col2 INTEGER
);
CREATE INDEX test_category_idx ON test2(category);
-- Added after edit to this question
CREATE INDEX test_id_idx ON test2(test1_id);
-- Populating with test data.
INSERT INTO test1(log_timestamp)
SELECT * FROM generate_series('2011-01-01'::timestamp, '2012-01-01'::timestamp, '1 hour');
INSERT INTO test2
SELECT id, substr(upper(md5(random()::TEXT)), 1, 1),
(20000*random()-10000)::int, (3000*random()-200)::int FROM test1;
INSERT INTO test2
SELECT id, substr(upper(md5(random()::TEXT)), 1, 1),
(2000*random()-1000)::int, (3000*random()-200)::int FROM test1;
INSERT INTO test2
SELECT id, substr(upper(md5(random()::TEXT)), 1, 1),
(2000*random()-40)::int, (3000*random()-200)::int FROM test1;
Here is a view that performs the most time-consuming operations:
CREATE VIEW testview1 AS
SELECT
t1.id,
t1.log_timestamp,
(SELECT SUM(t2.col1) FROM test2 t2 WHERE t2.test1_id=t1.id AND category='A') AS a,
(SELECT SUM(t2.col2) FROM test2 t2 WHERE t2.test1_id=t1.id AND category='B') AS b,
(SELECT SUM(t2.col1 - t2.col2) FROM test2 t2 WHERE t2.test1_id=t1.id AND category='C') AS c
FROM test1 t1;
SELECT a FROM testview1produces this plan (viaEXPLAIN ANALYZE):Seq Scan on test1 t1 (cost=0.00..1787086.55 rows=8761 width=4) (actual time=12.877..10517.575 rows=8761 loops=1) SubPlan 1 -> Aggregate (cost=203.96..203.97 rows=1 width=4) (actual time=1.193..1.193 rows=1 loops=8761) -> Bitmap Heap Scan on test2 t2 (cost=36.49..203.95 rows=1 width=4) (actual time=1.109..1.177 rows=0 loops=8761) Recheck Cond: ((category)::text = 'A'::text) Filter: (test1_id = $0) -> Bitmap Index Scan on test_category_idx (cost=0.00..36.49 rows=1631 width=0) (actual time=0.414..0.414 rows=1631 loops=8761) Index Cond: ((category)::text = 'A'::text) Total runtime: 10522.346 ms
SELECT a, a FROM testview1produces this plan:Seq Scan on test1 t1 (cost=0.00..3574037.50 rows=8761 width=4) (actual time=3.343..20550.817 rows=8761 loops=1) SubPlan 1 -> Aggregate (cost=203.96..203.97 rows=1 width=4) (actual time=1.183..1.183 rows=1 loops=8761) -> Bitmap Heap Scan on test2 t2 (cost=36.49..203.95 rows=1 width=4) (actual time=1.100..1.166 rows=0 loops=8761) Recheck Cond: ((category)::text = 'A'::text) Filter: (test1_id = $0) -> Bitmap Index Scan on test_category_idx (cost=0.00..36.49 rows=1631 width=0) (actual time=0.418..0.418 rows=1631 loops=8761) Index Cond: ((category)::text = 'A'::text) SubPlan 2 -> Aggregate (cost=203.96..203.97 rows=1 width=4) (actual time=1.154..1.154 rows=1 loops=8761) -> Bitmap Heap Scan on test2 t2 (cost=36.49..203.95 rows=1 width=4) (actual time=1.083..1.143 rows=0 loops=8761) Recheck Cond: ((category)::text = 'A'::text) Filter: (test1_id = $0) -> Bitmap Index Scan on test_category_idx (cost=0.00..36.49 rows=1631 width=0) (actual time=0.426..0.426 rows=1631 loops=8761) Index Cond: ((category)::text = 'A'::text) Total runtime: 20557.581 ms
Here, selecting a, a takes twice as long as selecting a, whereas they could really be computed just once. For example, with SELECT a, a+b, a-b FROM testview1, it goes through the sub-plan for a 3 times and through b twice, whereas the execution time could be reduced to 2/5 of the total time (assuming + and - are negligible here).
It's a good thing it doesn't compute the unused columns (b and c) when they're not needed, but is there a way to make it compute the same used columns from the view only once?
EDIT: @Frank Heikens correctly suggested to use an index, which was missing in the example above. While it does improve the speed for each sub-plan, it doesn't prevent the same sub-query to be computed multiple times. Sorry, I should have put this in the initial question to make it clear.
