Re: Is there any method to limit resource usage in PG? - Mailing list pgsql-general

From Igor Neyman
Subject Re: Is there any method to limit resource usage in PG?
Date
Msg-id A76B25F2823E954C9E45E32FA49D70EC1BBF8C17@mail.corp.perceptron.com
Whole thread Raw
In response to Is there any method to limit resource usage in PG?  (高健 <luckyjackgao@gmail.com>)
List pgsql-general
From: pgsql-general-owner@postgresql.org [mailto:pgsql-general-owner@postgresql.org] On Behalf Of ??
Sent: Monday, August 26, 2013 2:08 AM
To: pgsql-general
Subject: [GENERAL] Is there any method to limit resource usage in PG?

Hello:

Sorry for disturbing.

I am now encountering a serious problem: memory is not enough.

My customer reported that when they run a program they found the totall memory and disk i/o usage all reached to
thresholdvalue(80%).
 

That program is written by Java.
It is to use JDBC to pull out data from DB, while the query joined some table together,  It will return about  3000,000
records.
Then the program will use JDBC  again to write the records  row by row , to inert into another table in the DB.

My first question is:

Currently, my customer can not update there applications , and also they have no budget to buy new hardware. 

In fact, my customer want the program 's JDBC related part can be dealed with by PG at  this way:  
    The PG can control all of its process to run under usage  threshold.
    If  the resource will reach the threshold, then PG will slow processes down under the threshold.


I know that in Oracle, that there are resource plan etc. to approach this.
But in PG, I haven't find it.

And  I also want to know:

When there are so many records need to be  thrown to client, 
will PG use as much memory as it can?  
If  the total amount of records is larger than shared_buffer, will PG hung? or just do its work and communicate with
clientfor a few times till all the data is transferred?
 


Best Regards
  

No, PG does not have feature similar to Oracle's Resource Manager.

Regards,
Igor Neyman

pgsql-general by date:

Previous
From: Fujii Masao
Date:
Subject: Re: Re: postmaster.pid still exists after pacemaker stopped postgresql - how to remove
Next
From: Florian Weimer
Date:
Subject: Re: Problem creating index