Everything in MySQL seems to be done closer to the metal than in MSSQL, And the documentation treats it that way. Especially for optimization, you'll need to understand how indexes, system configuration, and the optimizer interact under various circumstances.
The "optimizer" is more a parser. In MSSQL your query plan is often a surprise (usually good, sometimes not). In MySQL, it pretty much does what you asked it to do, the way you expected it to. Which means you yourself need to have a deep understanding of the various ways it might be done.
Not built around a good TRANSACTION model (default MyISAM engine).
File-system setup is your problem.
All the database configuration is your problem - especially various cache sizes.
Sometimes it seems best to think of it as an ad-hoc, glorified isam. Codd and Date don't carry much weight here. They would say it with no embarrassment.