NAME
Class::DBI::Replication - Class::DBI for replicated database
SYNOPSIS
package Film;
use base qw(Class::DBI::Replication);
Film->set_master('dbi:mysql:host=master', $user, $pw);
Film->set_slaves(
[ 'dbi:mysql:host=slave1', $user, $pw ],
[ 'dbi:mysql:host=slave2', $user, $pw ],
);
DESCRIPTION
Classs::DBI::Replication extends Class::DBI's persistence for replicated
databases.
The idea is very simple. SELECT from slaves, INSERT/UPDATE/DELETE to
master.
From http://www.mysql.com/doc/R/e/Replication_FAQ.html,
Q: What should I do to prepare my client code to use
performance-enhancing replication?
A: If the part of your code that is responsible for database access
has been properly abstracted/modularized, converting it to run with
the replicated setup should be very smooth and easy - just change
the implementation of your database access to read from some slave
or the master, and to always write to the master.
With Class::DBI::Replication, it can be done easily!
METHODS
set_master
Film->set_master($datasource, $user, $password, \%attr);
This spcifies your master database. INSERT/UPDATE/DELETE are done
only to this database. Some SELECT queries also done to master for
concurrency problem.
If you don't want master to be distinct from slaves in SELECT
queries, put master in slaves, too.
set_slaves
Film->set_slaves(
[ 'dbi:mysql:host=slave1', $user, $password, \%attr ],
[ 'dbi:mysql:host=slave2', $user, $password, \%attr ],
);
This specifies your slave databases. SELECT are done to these
databases randomly. If you don't specify slaves, all queries are
gone to master, as always.
TODO
* More docs
* More testing
* retrieve() adter create() problem. Currently, SELECT calls inside
Class::DBI are done to master database.
* Concurrency problems
* Customizable slave picking algorithm like Round-Robin
AUTHOR
Tatsuhiko Miyagawa <miyagawa@bulknews.net>
This library is free software; you can redistribute it and/or modify it
under the same terms as Perl itself.
SEE ALSO
the Class::DBI manpage, the Class::DBI::mysql manpage