And then, you will have to get used to it. Meaning that there a lot of stuff that is so attractive, that you will want to explore the manual in search of guidance.
Let me give you a preview of what's there for you.
- Better start reading some theory on this stuff. If you have performance problems with huge loads of data, this is really what you should be doing. Partitioning is a chunk of pure performance hooked to a database. It's not in any way related to relational theory, and you don't need to redesign your database, but surely you need to rethink it in terms of partitioning.
Rush to read the manual and Robin Schumacher introductory article. If you really want to get the most about partitioning, though, you should check The Art of SQL (Note: I am not affiliated with the author or the publisher. This advice is out of my goodness only)
- Row based replication
- This is heavy stuff. For demanding application. If you have been dreaming of a more robust replication mechanism for MySQL, here it is.
Row based replication will send to the slave the data, rather than the statement to reproduce it. All edge cases where normal replication leave you in doubt are now handled efficiently by row-based replication. And if you want to get the best of the two worlds, you can still choose the mixed format. Check it out.
- Log tables on demand
- Get ready for a new experience. Logs whenever you need them, without scheduling a server reboot.
The new logs can materialize in your server in several shapes. You can have them as files, the same as you have seen them in previous versions, or you can have them as CVS tables (which you can convert to MyISAM or to something else), or you can have both at once, tables and files.
And your logs are dynamic. You can tell the server to start and stop logging, to change the log file name, to log on tables and files or only on one of them, you'll have more options than some video games.
And your life will change when you realize that you will need to plan for backups that include the logs. Yes. This is the downside. Get used to it.
- Event scheduler
- Someone said that the event scheduler is the coolest new feature in MySQL 5.1. Perhaps it was my reaction as well when I saw it for the first time. I can't tell you if it's the best one, but surely it will change your way of working. Say goodbye to OS dependent schedulers. Your database will take care of its own jobs. Setting up events is ridiculously easy. Once you try, you will be hooked forever.
- XML functions
- To tell you the truth, this is my least favorite feature in 5.1, because for theoretical reasons I am not fond of XML, but I have to admit that it's going to be damn useful, and I will have to get acquainted to it. It will save time and allow for flexible integration schemes with applications that only talk XML.