connection pooling in java

Hi,

We use connectionpooling in our webapplications. The connectionpooling is
setup in our application server. I understand that any webapplication in
that webserver can get and return connections from that pool.

Now we want to use this in our java-projects too but ... how is it
implemented ?

I mean, each application runs in a user's VM. When connections are
required, where's the pool from which to retrieve the connections?

Are these connection pools created inside each application and used by that
application only?
(like: AS400ConnectionPool appPool = new AS400ConnectionPool();


thanks for guiding me...

The connection pools are provided by the container so they're not
available for your stand-alone applications to use (directly).

Patrick Goovaerts wrote:[color=blue]
> Hi,
>
> We use connectionpooling in our webapplications. The connectionpooling is
> setup in our application server. I understand that any webapplication in
> that webserver can get and return connections from that pool.
>
> Now we want to use this in our java-projects too but ... how is it
> implemented ?
>
> I mean, each application runs in a user's VM. When connections are
> required, where's the pool from which to retrieve the connections?
>
> Are these connection pools created inside each application and used by that
> application only?
> (like: AS400ConnectionPool appPool = new AS400ConnectionPool();
>
>
> thanks for guiding me...
>
>[/color]

Not all connections pools are provided by a container, but I'd suspect
most that aren't container pools
(com.ibm.as400.access.AS400JDBCConnectionPool is an example) would only
pool connections within the JVM.

Now having written this, it seems like I heard or read about some JDBC
pooling done on a server for a whole set of clients. But it seems like
there could be a lot of issues to consider that would make the
single-JVM pooling adequate for most situations.

-Stevers

Patrick Goovaerts wrote:[color=blue]
> Hi,
>
> We use connectionpooling in our webapplications. The connectionpooling is
> setup in our application server. I understand that any webapplication in
> that webserver can get and return connections from that pool.
>
> Now we want to use this in our java-projects too but ... how is it
> implemented ?
>
> I mean, each application runs in a user's VM. When connections are
> required, where's the pool from which to retrieve the connections?
>
> Are these connection pools created inside each application and used by that
> application only?
> (like: AS400ConnectionPool appPool = new AS400ConnectionPool();
>
>
> thanks for guiding me...
>
>[/color]

meaning...

there are some possibilities to install a 'connectionpool' on a server which
can be accessed by all clients but considering the effort and problems which
can occur it is preferable to create a 'connectionpool' for each
application.
(in our case using: 'com.ibm.as400.access.AS400JDBCConnectionPool')
Needless to say that the connection pool is created only when heavy
data-access is needed.

This also means that the following is true also?
- "This also means that a user which loads 3 java-applications has also 3
connection pools active"
- "This can only avoided by attaching the application to eachother by
another application (menu) which can do the
connectionpooling for all apps attached. That way they are all running
inside 1 JVM."

thanks for your reply's.



"Steve Johnson-Evers" <severs@everbbbbrite.com> wrote in message
news:dgudhi$5tjg$1@news.boulder.ibm.com...[color=blue]
> Not all connections pools are provided by a container, but I'd suspect
> most that aren't container pools
> (com.ibm.as400.access.AS400JDBCConnectionPool is an example) would only
> pool connections within the JVM.
>
> Now having written this, it seems like I heard or read about some JDBC
> pooling done on a server for a whole set of clients. But it seems like
> there could be a lot of issues to consider that would make the
> single-JVM pooling adequate for most situations.
>
> -Stevers
>
> Patrick Goovaerts wrote:[color=green]
> > Hi,
> >
> > We use connectionpooling in our webapplications. The connectionpooling[/color][/color]
is[color=blue][color=green]
> > setup in our application server. I understand that any webapplication[/color][/color]
in[color=blue][color=green]
> > that webserver can get and return connections from that pool.
> >
> > Now we want to use this in our java-projects too but ... how is it
> > implemented ?
> >
> > I mean, each application runs in a user's VM. When connections are
> > required, where's the pool from which to retrieve the connections?
> >
> > Are these connection pools created inside each application and used by[/color][/color]
that[color=blue][color=green]
> > application only?
> > (like: AS400ConnectionPool appPool = new AS400ConnectionPool();
> >
> >
> > thanks for guiding me...
> >
> >[/color][/color]

Yes, yes and yes.

I think you're taking the approach that I would take.

-Stevers

Patrick Goovaerts wrote:[color=blue]
> meaning...
>
> there are some possibilities to install a 'connectionpool' on a server which
> can be accessed by all clients but considering the effort and problems which
> can occur it is preferable to create a 'connectionpool' for each
> application.
> (in our case using: 'com.ibm.as400.access.AS400JDBCConnectionPool')
> Needless to say that the connection pool is created only when heavy
> data-access is needed.
>
> This also means that the following is true also?
> - "This also means that a user which loads 3 java-applications has also 3
> connection pools active"
> - "This can only avoided by attaching the application to eachother by
> another application (menu) which can do the
> connectionpooling for all apps attached. That way they are all running
> inside 1 JVM."
>
> thanks for your reply's.
>
>
>
> "Steve Johnson-Evers" <severs@everbbbbrite.com> wrote in message
> news:dgudhi$5tjg$1@news.boulder.ibm.com...
>[color=green]
>>Not all connections pools are provided by a container, but I'd suspect
>>most that aren't container pools
>>(com.ibm.as400.access.AS400JDBCConnectionPool is an example) would only
>>pool connections within the JVM.
>>
>>Now having written this, it seems like I heard or read about some JDBC
>>pooling done on a server for a whole set of clients. But it seems like
>>there could be a lot of issues to consider that would make the
>>single-JVM pooling adequate for most situations.
>>
>>-Stevers
>>
>>Patrick Goovaerts wrote:
>>[color=darkred]
>>>Hi,
>>>
>>>We use connectionpooling in our webapplications. The connectionpooling[/color][/color]
>
> is
>[color=green][color=darkred]
>>>setup in our application server. I understand that any webapplication[/color][/color]
>
> in
>[color=green][color=darkred]
>>>that webserver can get and return connections from that pool.
>>>
>>>Now we want to use this in our java-projects too but ... how is it
>>>implemented ?
>>>
>>>I mean, each application runs in a user's VM. When connections are
>>>required, where's the pool from which to retrieve the connections?
>>>
>>>Are these connection pools created inside each application and used by[/color][/color]
>
> that
>[color=green][color=darkred]
>>>application only?
>>>(like: AS400ConnectionPool appPool = new AS400ConnectionPool();
>>>
>>>
>>>thanks for guiding me...
>>>
>>>[/color][/color]
>
>
>[/color]