Comment by Gergely Polonkai on "Is there any open source solution for database sharding ? Do I really want worry about it while I'm developing a web application ?"

Look at Vitess. They have awesome documentation, with a section dedicated on planning for scale.

If your engineering team knows how to operate Vitess, all the developers need to know is that. If your application is already complex enough, do a code review marathon and try to find anything that is against these “rules”.

Always keep your database schema under source control and provide unit test coverage of that schema. Also check schema changes into source control and run unit tests against the newly modified schema.
Think about code paths that can read from a replica vs. always choosing to read from the master. This will let you to scale your reads by just adding more replicas. Additionally, this will make it easy to expand into other data centers across the world.
Avoid complicated data relationships. Although RDBMS systems can handle them very well, such relationships hinder scaling in the future. When the time comes, it will be easier to shard the data. Avoid pushing too much logic into the database in the form of stored procedures, foreign keys, or triggers. Such operations overly tax the database and hinder scaling

(Excerpt from the site)

Search Hashnode