In response to "Deshpande, Yogesh Sadashiv (STSD-Openview)" <yogesh-sadashiv.deshpande@hp.com>:
> Hello ,
>
> We have a setup where in there are around 100 process running in parallel every 5 minutes and each one of them opens
aconnection to database. We are observing that for each connection , postgre also created on sub processes. We have set
max_connectionto 100. So the number of sub process in the system is close to 200 every 5 minutes. And because of this
weare seeing very high CPU usage.
This does not follow logically, in my experience. We have many servers that
have over 300 simultaneous connections, and the connections themselves do
not automatically create high CPU usage.
Unless of course, there is an issue with the particular OS you're using,
which you didn't mention.
> We need following information
>
> 1. Is there any configuration we do that would pool the connection request rather than coming out with
connectionlimit exceed.
Use pgpool or pgbouncer.
> 2. Is there any configuration we do that would limit the sub process to some value say 50 and any request for
connectionwould get queued.
Set the max connection and handle the connection retry in your application.
> Basically we wanted to limit the number of processes so that client code doesn't have to retry for unavailability for
connectionor sub processes , but postgre takes care of queuing?
pgpool and pgbouncer handle some of that, but I don't know if they do
exactly everything that you want. Probably a good place to start, though.
--
Bill Moran
http://www.potentialtech.com
http://people.collaborativefusion.com/~wmoran/
