How do I configure a MySQL Cluster with a redundant mysqld host?

Question

this is my first post here, so go easy on me.

I'm trying to rebuild my data architecture (which is currently a set of one primary MySQL server and a redundant fallback; the tables are primarily InnoDB for transactional features). This has worked well in the past, but with recent increases in traffic, it's not holding up very well.

At first I was considering adding a 'memory' cache table to speed things up, but I'd prefer something with transactions. So I've been looking into setting up a MySQL Cluster, but the documentation is a lot to take in, and I've now gotten myself very confused. Here's what I hope to accomplish:

I'd still like to have two mysqld machines, but I'd like to add two or three ndb nodes, and have the mysqld machines double as ndm-mgmd.

Is this a practical setup? How do I go about configuring this? If anyone knows of helpful examples or guides, I'd be pleased as punch (I've been searching, and each guide I stumble upon is twice as confusing as the last).

Answer 1

A great place to get started with MySQL Cluster are these Quick Start guides: http://mysql.com/products/cluster/get-started.html#quickstart

In terms of hardware profile, commodity servers are just fine - take a look at some of the recent benchmarks - 4.3m QPS across 8 x Intel dual socket servers with 24GB RAM: http://mysql.com/why-mysql/benchmarks/mysql-cluster/

Hope this helps. Let us know how you get along, and also go to the MySQL Cluster forum for direct access to engineering team: forums.mysql.com/list.php?25