Looping through an array of 2500 items would defiantly use some CPU, but that doesn't mean that it isn't the most efficient way to go. Could you provide just a top level overview of what the elements in the array actually are? Their purpose and origin would determine if a database is the way to go or not.

They've essentially search terms which have "perfect" results pages already created.
If people search for these things, there is a specific page they want to go to.

There is an Apache httpd method which involves a lookup text file or script:
http://httpd.apache.org/docs/2.2/mod/mod_r...html#rewritemap
http://httpd.apache.org/docs/2.2/rewrite/r...ernal-rewriting
But you need root access to the server.

If you want to cut down on CPU usage, you can do some basic indexing: if the first letter is "a", only look at the "a" array, (etc for each letter/number). But that assumes an exact match which is far more efficient than a partial match. If you want partial matching, I think a database lookup (using LIKE on an indexed table) would be the way to go. But an index again will only work best for something like "abcd%" and not "%abcd%". It really depends on what sort of efficiency you need. If you can use some indexing it will make a huge difference, database or not.

I'm only looking for exact matches, so I'll index them by first letter and see how that goes.
That, however, is work for another day. I'll try and get back to you with my results.

PHP in_array() is probably the quickest way that I know of.