Hi Stu,

There is many good pooling framework, and HikariCP is one of the best of them.

This framework is stable, has a good community, is light (130kb - c3p0 487kb - dbcp2 171kb) and code is clean (c3p0 is unreadable).

A pool must present a Connection instance normally indistinguishable from a "fresh" Connection. I recommand reading https://github.com/brettwooldridge/HikariCP/wiki/%22My-benchmark-doesn't-show-a-difference.

HikariCP does the job well and it's fast.

The example with HilkariCP was written just to have one example project (for support), there is no known caveat using HikariCP with the MariaDB and there is big companies using the combination in production.

Diego.

On Sun, Apr 17, 2016 at 5:20 PM, Colin Charles <byte@aeon.com.my> wrote:

Hi!

I think Kristian answered the questions about searching (I too recommend using Google search)

> On 16 Apr 2016, at 09:38, Stu Smith <stu26code@gmail.com> wrote:
>
> I noticed the MariaDB connector/J docs give an example using the HikariCP connection pool. Was there any reason for this?
>
> I saw they filed bugs in the mariadb jira, so they have put some effort into working with the community - and they have pretty graphs showing how fast they are.
>
> We're not really too worried about optimizing it yet - just adding a connection pool as a best practice kind of thing to a rest service hitting mariadb. Just wanted to make sure there weren't any caveats to using it with connector/J or if it was picked as the example for any particular reason.

As for the above, I’ve added wlad and diego whom may be able to hep you more on Connector/J

cheers,
-colin

--
Colin Charles, http://bytebot.net/blog/
twitter: @bytebot | skype: colincharles
"First they ignore you, then they laugh at you, then they fight you, then you win." -- Mohandas Gandhi