AWS MySQL Hight availability, what is the best way to do it?

Question

MySQL

Greetings. The task is to raise a web server with MySQL Master on Amazon (the slaves will be located in other DCs), you need to ensure the constant performance of MySQL and the server as a whole, how best to do this?

PS So far, I have subtracted for MySQL this AWS RDS and MySQL NDBCluster
AWS RDS as I imagine - this is not controlled by mysql server, i.e. I can't optimize it as I need or am I wrong?
MySQL NDBCluster - it has a point of failure in the form of a "managing" MySQL

PSS And another question arose: EBS can any FS?

Thank you.

Answer 1

In what "other DCs"? It is better not to mess with NDB. In RDS, some of the settings are available to you, some are not. If I were you, I would look at Scalr or Galera.
EBS are just virtual disks, devices. Naturally, they can be formatted however you like.

Answer 2

In RDS, I have not seen options for master master replication, there is quite a lot of things available for optimization, at least everything that I can optimize in my.cnf I can optimize through group in rds.
MySQL NDBCluster - it has a point of failure in the form of a "managing" MySQL
There is no such point of failure there.