I'm considering replication with MySQL as a way to improve concurrency on our application but am wondering how I go about selecting a slave database for my reads?

I should say this is a web app, and ideally when a session is created I want it to stay with a particular server (this is based on my understanding that not all slaves will necessary be bang up to date depending on load etc?).

I guess I have 3 solutions:

1. pick a slave in my application code - I could do this, but it doesn't seem very intelligent as it may not evenly spread the load.

2. some sort of mysql configuration that sees all slaves as a single server? Don't know if this is possible or how it works in practise.

3. a hardware load balancer - don't really know anything abou this, but presumably solutions exist?

If anyone could shed any light on any of the above I'd greatly appreciate it.

Thanks
Tom

Based on your req.. u need a load balancer. Yes you can use this replications to meet your goals.
But if you are using autoincrement ids, then you does not solve your problems.
Try this

http://www.onlamp.com/pub/a/onlamp/2...plication.html