Magento Database Scaling?

I've been doing some work on Wordpress and just stumbled across database scaling.

I开发者_StackOverflow was wondering if it was at all possible to implement this with Magento as we have an increasingly large database and thousands of products still to add (with various options for configurable products!!)

You can setup master slave replication for mysql so that you have a customer facing server on the master database and a admin only server on the slave database. You can edit /app/etc/local.xml on the slave so that it reads from the slave but writes to the master. In that way you can upload loads of products, do loads of reports and the live system only has to slow down for the occasional write to the db. You can also have your slave system in your office on the end of an ADSL line. For this to work standard mysql master slave settings work fine, it also does rather well in catching up with itself if the slave needs rebooting. The only gotcha of note is that mysql log files get big and your mysql my.cnf needs to have the setting on to delete the log files after n days.

Magento has recently certified ClustrixDB to be the scale-out database solution: http://www.marketwired.com/press-release/clustrix-selected-as-magento-technology-partner-2010010.htm

Disclaimer: I work at Clustrix and I know this is just a press release, but I have been also involved in the testing and it really works!

Disclaimer - I work at Clustrix.

Magento solution architect and certified developer, Kevin Bortnick, recently gave a webinar entiteld "Scaling Techniques to Increase Magento Capacity" at Meet Magento NY. Kevin currently works for clustrix but prior to joining this year, he worked on large Magento sites such as The Home Depot and Dr. Jays. In the webinar he shares his practical experiences. He covers 3 front end strategies and 8 RDBMS scaling strategies including:

1. Percona
2. Faster hardware
3. Master with read slaves
4. Master/Master, Store/Admin
5. True Multi-master
6. Partitioning (Magento 2 feature CQRS)
7. NoSQL
8. ClustrixDB

I hope this webinar helps you find the right solution for your use case.