Обсуждение: Slow Query - Postgres 9.2
Hi all...
I'm working on a Slow Query. It's faster now (It was 20sec before) but still not good.
Can you have a look and see if you can find something?
Cheers
Query:
WITH jobs AS ( SELECT job.id, job.clientid, CONCAT(customer.company, ' ', customer.name_first, ' ', customer.name_last) AS "identity", job.gps_lat, job.gps_long FROM public.ja_jobs AS job JOIN public.ja_customers AS customer ON customer.id = job.customerid WHERE job.clientid = 22 AND job.time_job >= 1422702000 AND job.time_job <= 1456743540
AND NOT job.deleted AND NOT job.templated
), items AS ( SELECT job.identity, COALESCE(bill_item.unit_cost, billable.unit_cost, 0) * bill_item.quantity AS cost, COALESCE(bill_item.unit_price, billable.unit_price, 0) * bill_item.quantity AS revenue, bill_item.quantity AS quantity, note.n_quote_status, bill_item.for_invoicing FROM jobs AS job JOIN public.ja_notes AS note ON note.jobid = job.id AND note.note_type IN ('time', 'part') JOIN dm.bill_items AS bill_item ON bill_item.bill_item_id = note.bill_item_id AND bill_item.for_invoicing LEFT JOIN dm.billables AS billable ON billable.billable_id = note.billable_id JOIN public.ja_mobiusers AS user_creator ON user_creator.id = note.mobiuserid AND ( user_creator.name_first ilike 'Alan' OR user_creator.name_last ilike 'Alan' )
)
SELECT item.identity, SUM(CASE WHEN item.for_invoicing THEN item.revenue ELSE 0 END) AS revenue, SUM(CASE WHEN item.for_invoicing THEN item.quantity ELSE 0 END) AS quantity, SUM(CASE WHEN item.for_invoicing THEN item.cost ELSE 0 END) AS cost
FROM items AS item
GROUP BY item.identity
ORDER BY revenue DESC, item.identity ASCExplain analyze link: http://explain.depesz.com/s/IIDj
On 3/2/16, drum.lucas@gmail.com <drum.lucas@gmail.com> wrote:
> Hi all...
>
> I'm working on a Slow Query. It's faster now (It was 20sec before) but
> still not good.
>
> Can you have a look and see if you can find something?
> Cheers
>
> Query:
>
> WITH jobs AS (
> ...
> FROM
> jobs AS job
> JOIN
> public.ja_notes AS note
> ON
> note.jobid = job.id
> AND note.note_type IN ('time', 'part')
> ...
It is the most long part. All query is 8.8sec.
SeqScan by CTE is 2.8sec! and index scan in ix_notes_jobid_per_type
500rows(loops) * 9.878ms!!! = 4.939sec.
Why does it take so long time?
For example, index scan in ja_customers_pkey is only 0.781 per row...
10 times faster!
What definition of the ix_notes_jobid_per_type? Is it bloated?
> Explain analyze link: http://explain.depesz.com/s/IIDj
>
--
Best regards,
Vitaly Burovoy
On 3 March 2016 at 10:33, Vitaly Burovoy <vitaly.burovoy@gmail.com> wrote:
On 3/2/16, drum.lucas@gmail.com <drum.lucas@gmail.com> wrote:
> Hi all...
>
> I'm working on a Slow Query. It's faster now (It was 20sec before) but
> still not good.
>
> Can you have a look and see if you can find something?
> Cheers
>
> Query:
>
> WITH jobs AS (
> ...
> FROM
> jobs AS job
> JOIN
> public.ja_notes AS note
> ON
> note.jobid = job.id
> AND note.note_type IN ('time', 'part')
> ...
It is the most long part. All query is 8.8sec.
SeqScan by CTE is 2.8sec! and index scan in ix_notes_jobid_per_type
500rows(loops) * 9.878ms!!! = 4.939sec.
Why does it take so long time?
For example, index scan in ja_customers_pkey is only 0.781 per row...
10 times faster!
What definition of the ix_notes_jobid_per_type? Is it bloated?
Hi there!
CREATE INDEX
ix_notes_jobid_per_type
ON
ja_notes
(
"jobid",
"note_type"
);
\di+ ix_notes_jobid_per_type
List of relations
Schema | Name | Type | Owner | Table | Size | Description
--------+-------------------------+-------+----------+----------+--------+-------------
public | ix_notes_jobid_per_type | index | postgres | ja_notes | 484 MB |
it does not seem to be bloated... since the table is 2805 MB
On 3/2/16, drum.lucas@gmail.com <drum.lucas@gmail.com> wrote:
> On 3 March 2016 at 10:33, Vitaly Burovoy <vitaly.burovoy@gmail.com> wrote:
>> On 3/2/16, drum.lucas@gmail.com <drum.lucas@gmail.com> wrote:
>> > Hi all...
>> >
>> > I'm working on a Slow Query. It's faster now (It was 20sec before) but
>> > still not good.
>> >
>> > Can you have a look and see if you can find something?
>> > Cheers
>> >
>> > Query:
>> >
>> > WITH jobs AS (
>> > ...
>> > FROM
>> > jobs AS job
>> > JOIN
>> > public.ja_notes AS note
>> > ON
>> > note.jobid = job.id
>> > AND note.note_type IN ('time', 'part')
>> > ...
>>
>> It is the most long part. All query is 8.8sec.
>> SeqScan by CTE is 2.8sec! and index scan in ix_notes_jobid_per_type
>> 500rows(loops) * 9.878ms!!! = 4.939sec.
>>
>> Why does it take so long time?
>> For example, index scan in ja_customers_pkey is only 0.781 per row...
>> 10 times faster!
>>
>> What definition of the ix_notes_jobid_per_type? Is it bloated?
>>
>>
>> Hi there!
>
> CREATE INDEX
> ix_notes_jobid_per_type
> ON
> ja_notes
> (
> "jobid",
> "note_type"
> );
>
>
> \di+ ix_notes_jobid_per_type
>
> List of relations
>
> Schema | Name | Type | Owner | Table | Size |
> Description
>
> --------+-------------------------+-------+----------+----------+--------+-------------
>
> public | ix_notes_jobid_per_type | index | postgres | ja_notes | 484 MB |
>
> it does not seem to be bloated... since the table is 2805 MB
I'm sorry for the late answer.
It doesn't seem there is something wrong, but in the row:
> Index Scan using "ix_notes_jobid_per_type" on "ja_notes" "note"
> (cost=0.00..29.33 rows=6 width=34)
> (actual time=8.530..9.878 rows=1 loops=500)
the first digit is the time when the first row was returned from. See
[1] for more details.
It looks very strange for me because IIUC it spends almost all time on
opening (reopening) the index.
For example, <Index Scan using "ja_customers_pkey"> has an "actual
time" with both values equal.
I'd try to reindex "ix_notes_jobid_per_type".
[1]http://www.depesz.com/2013/04/16/explaining-the-unexplainable/
--
Best regards,
Vitaly Burovoy