I am beginning a website which depends on search. While it's most likely likely to search fundamental meta data at first, it could grow to something bigger later on.

So which DB/DB Engine is better inside your opinion if this involves search performance and future scalability?

Thank you for help

It is dependent on what you're searching.

Should you doing lots of text searching then you definitely want not only a database - additionally you desire a search formula. You'll find them online plus they may use several databases as backends.

However, should you only simple text searches then MySQL MyISAM offers full-text searching that we use for small quantities of text (under a couple of GB).

Other searches include using secrets and indexes that might make you PostgreSQL for it's reliable Acidity compliance or MySQL with INNODB.

What's "search" ? What exactly are you searching for and what type of queries would you expect?

PostgreSQL is extremely powerfull, has full text search, btree, hash, gin and gist indexes. You may also configure you have types and operators, things are there to optimize your searches within the database. The choice is yours to make use of and tweak it for you personally situation.

PostgreSQL is simple to use with PHP, not a problem whatsoever. And it is free, kind of BSD-licence.

Based on anything you mean by “search” any database system might work. (MySQL is really a well know and fast RDBMS).

If what you're searching for is really “full text search” then you definitely should have a look into MySQL FULLTEXT indices (only functional using the MyISAM after sales, IIRC), Lucene or Xapian.

The Zend Framework (designed in PHP) includes a ready adapter for lucene, see: http://devzone.zend.com/article/91