MySQL is
only competitive with single table queries - which rarely happen in real life.
There's an awful lot of BS floating around about MySQL, much of it initiated by MySQL AB. Basically, it's ideal as a back end for a web server ... and that's about it. It is slow at any sort of realistic query, collapses under load, and is poor at handling updates.
More importantly, it just lacks the ability to perform any kind of operation that goes much beyond a simple join. I've seen its SQL support described as a much less capable version of that available in Microsoft Access - hardly a ringing endorsement.
To be fair, the trend is to move business rules into the middleware, which takes a lot of pressure off the back-end. But I haven't heard of many people trying to build 'real' n-tier products with MySQL; it's just too primitive.
MaxDB is
not MySQL. It is a completely separate product that MySQL AB has gained rights to. It is also charged per seat, albeit at a rate about one quarter that of Oracle. And the support costs would make Oracle Corp blush.
I'm cynical enough to suggest that its performance is likely to be proportionate to its cost, when compared to Oracle or DB2. And I've no doubt that development costs will be higher.
A great deal of public support for MySQL is badly misguided. There are other open source solutions that are far, far more competent and offer competitive performance. PostgreSQL and Firebird spring to mind immediately. The MySQL bandwagon is sucking interest away from these worthy products, and in the event of a MySQL AB trainwreck, will probably tarnish the whole idea of seeking alternatives to Oracle and MS SQL Server.