Replication implmentation

[Aiya]

Hello everyone! hope that here i can find a help, which i really needed.
actually i am new to this topic that's why i completely stucked.
The situation is that i know how replication algorithms works theoretically, but i have no idea how to implemet it practically. I searched lots of information, and didn't find definite explanation about implementation.
The thing is that i have Darwin cluster consists of 16 nodes, running uder Unix. Suppose that i want to replicate files in this 3 nodes, so here is the question: what is the system structure should be?
or, for example, i will have the simple banking system program written on C, the data will be stored in the files, so I have to replicate this files on these nodes. Also i will write code for the primary backup replication.
Then what should i have to do? Or how to organise the system? How to run the programms? do i need some framework?
Please help me, maybe my question is too vague, but i've try to reflect the general idea of my problem.
Thanks a lot.

[Candace]

Here is my example. I have 2 servers & MySQL database. One (server_client) hosts our database client and the other (server_host) hosts the database provider. Server_client We want to replicate the database supplier of server_host and then replicates the basic customer server_client. Before you start to configure our servers, they must be given a unique identifier to coexist several replications on th network. This identifier is specified in the configuration file using the directive: server-id