Postgresql – How to increase performance for select in select statement

performancepostgresqlquery-performance

The item table is like this

id    category_id    name    registered

I have to show total number of items from the beginning. Suppose between 2013-05-01 to 20-05–05, I have to count all the items before 2013-05-01 and add that to count of 2013-05-01 's items. The result have to be grouped by registered and category_id. I have to do this in one query. This is my query

SELECT registered, me.category_id, 
    COUNT( * ) + (select count(*) from items where me.category_id = category_id 
    and registered < '2013-05-01 00:00:00') AS Value FROM items me

WHERE (
    (
      category_id IS NOT NULL
      AND 
     (me.registered BETWEEN '2013-05-01 00:00:00' AND '2013-05-02 00:00:00') 
    ) 
) 
GROUP BY registered, category_id

It takes almost 28 secs to generate the result. How can i increase the performance.

Update
I am using pgSQL. The registered and category_id is indexed. There are 568814 rows. I will be also very grateful, if any one can tell me that my sql is correct or not.

Update
Sorry for late reply.

CREATE TABLE items
(
id uuid NOT NULL DEFAULT uuid_generate_v4(),
registered timestamp without time zone NOT NULL,
modified timestamp without time zone NOT NULL,
published timestamp without time zone,
owner_id uuid,
status_id integer NOT NULL,
category_id uuid NOT NULL,
location_id uuid,
title text,
description text,
quantity integer NOT NULL DEFAULT 1,
unit text,
price_low bigint,
price bigint,
currency text NOT NULL DEFAULT 'BDT'::text,
ip text,
attributes_json text,
from_mobile boolean NOT NULL DEFAULT false,
CONSTRAINT items_pkey PRIMARY KEY (id),
CONSTRAINT items_fk_category_id FOREIGN KEY (category_id)
        REFERENCES categories (id) MATCH SIMPLE
        ON UPDATE CASCADE ON DELETE CASCADE DEFERRABLE INITIALLY IMMEDIATE,
CONSTRAINT items_fk_location_id FOREIGN KEY (location_id)
        REFERENCES locations (id) MATCH SIMPLE
        ON UPDATE NO ACTION ON DELETE NO ACTION DEFERRABLE INITIALLY IMMEDIATE,
CONSTRAINT items_fk_owner_id FOREIGN KEY (owner_id)
        REFERENCES users (id) MATCH SIMPLE
        ON UPDATE CASCADE ON DELETE CASCADE DEFERRABLE INITIALLY IMMEDIATE,
CONSTRAINT items_fk_status_id FOREIGN KEY (status_id)
        REFERENCES item_statuses (id) MATCH SIMPLE
      ON UPDATE CASCADE ON DELETE CASCADE DEFERRABLE INITIALLY IMMEDIATE
)

Best Answer

What kind of column is [registered] -> datetime or varchar
Why are you grouping by the full value of the [registered] column ?
Ex. If you have the following values for this column:
```
    2013-05-01 00:00:00
    2013-05-01 00:00:01
    2013-05-01 00:00:02
    2013-05-01 00:00:03
    2013-05-01 00:00:04
```
... and you group by them, you will have 5 rows in the output.

Grouping at the minute or hour level will get you less results -> faster query execution time ... if you have 568.814 rows, how many rows does your query return ? I suppose the result set is very close to 568.814 rows :)
In the current form, for each row generated by the primary query, the subquery has to be run. Try to avoid that ...

Here my version of your query. Based on the same input, the output will be the same:

SELECT
  t1.registered,
  t1.category_id,
  (COUNT(*) + COALESCE(q2.c2,0)) AS Value
FROM
  items t1
  LEFT JOIN
  (
    SELECT
      t2.category_id,
      COUNT(*) AS c2
    FROM
      items t2
    WHERE 
      t2.registered < '2013-05-01 00:00:00'
    GROUP BY
      t2.category_id
  ) q2
  ON  t1.category_id = q2.category_id
WHERE
      t1.registered BETWEEN '2013-05-01 00:00:00' 
                        AND '2013-05-02 00:00:00'
GROUP BY
  t1.registered,
  t1.category_id,
  q2.c2 ;

Output

REGISTERED              CATEGORY_ID     VALUE
---------------------------------------------
May, 01 2013 00:00:00+0000  1           5
May, 01 2013 01:00:00+0000  1           5
May, 01 2013 02:00:00+0000  2           6
May, 01 2013 07:00:00+0000  3           3
May, 01 2013 08:00:00+0000  5           3
May, 01 2013 09:00:00+0000  1           5

Query Plan

HashAggregate (cost=35.27..35.32 rows=4 width=20)
-> Hash Right Join (cost=33.63..35.23 rows=4 width=20)
Hash Cond: (t2.category_id = t1.category_id)
-> HashAggregate (cost=20.82..21.49 rows=67 width=4)
-> Bitmap Heap Scan on items t2 (cost=6.27..19.52 rows=260 width=4)
Recheck Cond: (registered < '2013-05-01 00:00:00'::timestamp without time zone)
-> Bitmap Index Scan on idx_reg (cost=0.00..6.20 rows=260 width=0)
Index Cond: (registered < '2013-05-01 00:00:00'::timestamp without time zone)
-> Hash (cost=12.76..12.76 rows=4 width=12)
-> Bitmap Heap Scan on items t1 (cost=4.29..12.76 rows=4 width=12)
Recheck Cond: ((registered >= '2013-05-01 00:00:00'::timestamp without time zone) AND (registered <= '2013-05-02 00:00:00'::timestamp without time zone))
-> Bitmap Index Scan on idx_reg (cost=0.00..4.29 rows=4 width=0)
Index Cond: ((registered >= '2013-05-01 00:00:00'::timestamp without time zone) AND (registered <= '2013-05-02 00:00:00'::timestamp without time zone))

For anyone interested, here is the sql fiddle page with sample data: SQL-Fiddle-2

Related Solutions

Postgresql – Postgres multiple joins slow query, how to store default child record

You write:

Each customer can have multiple sites, but only one should be displayed in this list.

Yet, your query retrieves all rows. That would be a point to optimize. But you also do not define which site is to be picked.

Either way, it does not matter much here. Your EXPLAIN shows only 5026 rows for the site scan (5018 for the customer scan). So hardly any customer actually has more than one site. Did you ANALYZE your tables before running EXPLAIN?

From the numbers I see in your EXPLAIN, indexes will give you nothing for this query. Sequential table scans will be the fastest possible way. Half a second is rather slow for 5000 rows, though. Maybe your database needs some general performance tuning?

Maybe the query itself is faster, but "half a second" includes network transfer? EXPLAIN ANALYZE would tell us more.

If this query is your bottleneck, I would suggest you implement a materialized view.

After you provided more information I find that my diagnosis pretty much holds.

The query itself needs 27 ms. Not much of a problem there. "Half a second" was the kind of misunderstanding I had suspected. The slow part is the network transfer (plus ssh encoding / decoding, possibly rendering). You should only retrieve 100 rows, that would solve most of it, even if it means to execute the whole query every time.

If you go the route with a materialized view like I proposed you could add a serial number without gaps to the table plus index on it - by adding a column row_number() OVER (<your sort citeria here>) AS mv_id.

Then you can query:

SELECT *
FROM   materialized_view
WHERE  mv_id >= 2700
AND    mv_id <  2800;

This will perform very fast. LIMIT / OFFSET cannot compete, that needs to compute the whole table before it can sort and pick 100 rows.

pgAdmin timing

When you execute a query from the query tool, the message pane shows something like:

Total query runtime: 62 ms.

And the status line shows the same time. I quote pgAdmin help about that:

The status line will show how long the last query took to complete. If a dataset was returned, not only the elapsed time for server execution is displayed, but also the time to retrieve the data from the server to the Data Output page.

If you want to see the time on the server you need to use SQL EXPLAIN ANALYZE or the built in Shift + F7keyboard shortcut or Query -> Explain analyze. Then, at the bottom of the explain output you get something like this:

Total runtime: 0.269 ms

Postgresql – Improve performance on concurrent UPDATEs for a timestamp column in Postgres

Would a different kind of column be faster? For example an integer

No. timestamp and timestamptz are just unsigned 64-bit integers internally anyway.

Is there some way to not lock the column?

It doesn't lock the column. It takes weak table lock that doesn't really block anything except DDL, and takes a row level lock on the row you're updating.

There is no way to prevent the row level lock. It exists because without it behaviour and ordering concurrent updates would be undefined. We don't like undefined behaviour in RDBMSs.

It only blocks concurrent updates of the same row anyway.

Any other tips to improve this?

Not with the detail provided. There's likely a better way to do what you're trying to do, but it'll probably involve taking a few steps back and looking for a different strategy for solving the underlying problem.

In the specific case of cache invalidation I think you might want to look into LISTEN and NOTIFY. Again though, there just isn't enough info here to go on.

Best Answer

Related Solutions

Postgresql – Postgres multiple joins slow query, how to store default child record

pgAdmin timing

Postgresql – Improve performance on concurrent UPDATEs for a timestamp column in Postgres

Related Question