Le 08/10/2015 01:40, Carlo a écrit :
>> How many cores do you have on that machine?
Test if limiting number of simultaneous feeds, like bringing their
number down to half of your normal connections has the same positive
effect.
<<
I am told 32 cores on a LINUX VM. The operators have tried
Le Mardi 21 Octobre 2014 10:44 CEST, David Rowley a
écrit:
> For what it's worth I'd say they are identical, at least, if you discount
> deferring foreign key constraints or also executing the query from within
> a volatile function which was called by a query which just updated the
> user_inf
Le 20/10/2014 15:58, Tom Lane a écrit :
Laurent Martelli writes:
Do we agree that both queries are identical ?
No, they *aren't* identical. Go consult any SQL reference. Left join
conditions don't work the way you seem to be thinking: after the join,
the RHS column might be nu
d is not null
does not use any *new* information from user_user_info.
Regards,
Laurent
Le 19/10/2014 10:41, David Rowley a écrit :
On Sun, Oct 19, 2014 at 5:10 PM, Laurent Martelli
mailto:laurent.marte...@enercoop.org>>
wrote:
Hello there,
I have a strange query plan involv
Hello there,
I have a strange query plan involving an IS NOT NULL and a LEFT JOIN.
I grant you that the query can be written without the JOIN on
user_user_info,
but it is generated like this by hibernate. Just changing the IS NOT
NULL condition
to the other side of useless JOIN makes a big di
ures.
I just wished there was a means to fully automate all this and render
it transparent to the user, just like an index.
Merlin> Voila! Merlin p.s. normalize your data always!
I have this:
pictures(
PictureID serial PRIMARY KEY,
Owner integer NOT NULL REFERENCES users,
[...]);
CREATE
4.38 rows=21 width=4) (actual time=7.585..7.605
rows=21 loops=1)
-> Seq Scan on pictures (cost=0.00..103.70 rows=4270 width=4) (actual
time=0.015..3.272 rows=4270 loops=1)
Total runtime: 7.719 ms
--
Laurent Martelli
[EMAIL PROTECTED]
>>>>> "Rod" == Rod Taylor <[EMAIL PROTECTED]> writes:
Rod> On Wed, 2005-03-16 at 18:58 +0100, Laurent Martelli wrote:
>> Consider this query:
>>
>> SELECT distinct owner from pictures;
Rod> The performance has nothing to do
for 20 rows. I looked at other type of indexes,
but they seem to either not give beter perfs or be irrelevant.
Any ideas, apart from more or less manually maintaining a list of
distinct owners in another table ?
--
Laurent Martelli
[EMAIL PROTECTED]Jav
tient.
alix> Pour moi la même chose que l'année dernière : on y va.
Idem pour moi.
--
Laurent Martellivice-président de Parinux
http://www.bearteam.org/~laurent/ http://www.parinux.org/
[EMAIL PROTECTED]
---
avoir qq infos et des prix.
Je pense que c'est un bon investissement.
+1 donc
--
Laurent Martellivice-président de Parinux
http://www.bearteam.org/~laurent/ http://www.parinux.org/
[EMAIL PROTECTED]
---(en
é d'occasion sans facture et je
Laurent> vois mal une assurance accepter de rembourser un appareil
Laurent> sans cet élément.
De mémoire, le gars nous avait filé la facture lorsqu'il nous l'a vendu.
--
Laurent Martellivice
odules)
Laurent> Et subversion ?
Puisqu'on a pas d'existant sous CVS à migrer, je suis aussi plutôt
commencer directement avec subversion, qui est un CVS en mieux.
--
Laurent Martellivice-président de Parinux
http://www.bearteam.org/~laurent
How in hell did could this mail be sent to pgsql-performance ??? I
must have inadvertently hit a fatal and obscure keystroke in
Emacs/Gnus.
Sorry for the noise.
--
Laurent Martelli
[EMAIL PROTECTED]Java Aspect Components
http://www.aopsys.com
anifestation maitenant et tu peux pas dire non ;)
Ouf! Je suis soulagé de la tournure que ça prends.
--
Laurent Martellivice-président de Parinux
http://www.bearteam.org/~laurent/ http://www.parinux.org/
[EMAIL PROTECTED]
>>>>> "Tom" == Tom Lane <[EMAIL PROTECTED]> writes:
Tom> Laurent Martelli <[EMAIL PROTECTED]> writes:
>> Now, if I use the following view to abstract access rights:
>> CREATE VIEW userpictures (
>> PictureID,RollID,FrameID,Des
Filter: ((value)::text = 'laurent'::text)
-> Hash (cost=5.19..5.19 rows=12 width=8) (actual time=0.198..0.198
rows=0 loops=1)
-> Seq Scan on groupsdef (cost=0.00..5.19 rows=12 width=8)
(actual time=0.031..0.178 rows=11 loops=1)
Filter: (userid = 2)
Total runtime: 35.657 ms
--
Laurent Martelli
[EMAIL PROTECTED]Java Aspect Components
http://www.aopsys.com/ http://jac.objectweb.org
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
>>>>> "Tom" == Tom Lane <[EMAIL PROTECTED]> writes:
Tom> Laurent Martelli <[EMAIL PROTECTED]> writes:
>> The pictures table is scanned, but it's not needed.
Tom> Yes it is. For example, if pictures is empty then the view
Tom> yiel
(cost=100.28..100.37 rows=38 width=4) (actual time=0.091..0.094
rows=8 loops=1)
Sort Key: topicscontent.pictureid
-> Index Scan using topicscontent_topicid on topicscontent
(cost=0.00..99.28 rows=38 width=4) (actual time=0.039..0.057 rows=8 loops
Shridhar> classes.id=lists.value::integer.
With classes.id of type integer and lists.value of type varchar, I get
"ERROR: Cannot cast type character varying to integer", which is not
such a surprise.
Thanks for your help anyway.
--
Laurent Martelli
[EMAIL PROTECTED]
>>>>> "Shridhar" == Shridhar Daithankar <[EMAIL PROTECTED]> writes:
Shridhar> Laurent Martelli wrote:
>>>>>>> "Shridhar" == Shridhar Daithankar
>>>>>>> <[EMAIL PROTECTED]> writes:
Shridhar&g
>>>>> "Shridhar" == Shridhar Daithankar <[EMAIL PROTECTED]> writes:
Shridhar> Laurent Martelli wrote:
[...]
>> Should I understand that a join on incompatible types (such as
>> integer and varchar) may lead to bad performances ?
Shridh
"inner".id)
scott> This estimate is WAY off. Are both of those fields indexed
scott> and analyzed? Have you tried upping the statistics target on
scott> those two fields? I assume they are compatible types.
Should I understand that a join on incompatible types (such as i
23 matches
Mail list logo
