1) Do you use a load-balancer for your MySQL Deployment?
Yes. Well, up till just a little while ago...
2) What load-balancer (product name) do you use for your MySQL Deployment?
We used to run MySQL's NDB Cluster behind a hardware load balancer (don't remember the name, but it's not that relevant).
3) Do you use the default mechanisms of the load-balancer to negotiate traffic to your MySQL deployment, or have you created your own custom mechanism for the load-balancer to use?
Default.
4) (a) Other than your current load-balancer, have you tried to use any other load-balancers with success or failure? (b) Or is there another load-balancer you are looking into possibly using? (Some Examples: MySQL Proxy, Continuent, Sequoia (C-JDBC), Linux Virtual Server, F5 Networks BigIP, EddieDNS, or even Heartbeat, Pen, Python Director, Distributor)
We tried Continuent about two years ago and had bad experiences with it. It was a three node setup and as long as we didn't write to it it was fine, but otherwise it would regularly hang. Frequently we had to reboot everything to get it unstuck. We never put it into production.
But that was two years ago and I dont know anything about it since.
5) How do you primarily use the load balancer? 1) load-balance read-only SQL queries 2) load-balance read-write SQL queries 3) other? (like some custom setup)
2.
6) When it comes to scaling MySQL and the use of load-balancing, what do you feel is a technology that is missing that the MySQL community should create? (I.e. perhaps some new technical item in the MySQL database server software, or something on the load-balancing technology side)
Something similar to pgpool that can automatically redirect writes to the master and reads to a pool of slaves.
http://pgpool.projects.postgresql.org/ If that already exists and you know about it, link please! :) -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
