CareerCup

Map-Reduce....(Since the question does not mention a limit on resources...)

assume 1 web page is 18KB (keep the math simple)

n billion web page is about n x 10^9 x 18 x 10 ^3 B = 18n x 10^12 B
30 min is 1800 seconds

18nx10^12B /1800seconds = 10n x 10^9 B/s = 80n x 10^9 b/s = 80n Gbit/s

If selecting 2n x 40Gbit/s network processor with built-in DPI, one can solve this problem.

The numbers do have sematic meaning, eg: location. Hence, if the database is distributed, that helps. Additionally, the DB must be stored in a N-nary search tree where N is large, reducing the height of the tree significantly.

Sorry, the question is not clear. Could you please provide some context?

Regular expressions. Use something like "ls -aR | grep <pattern>".

He asked my...that I have one single large DB and I have just 30 min to search it. If I use a regular expression to search a DB like that it would take me a time greater that 30 min.

Using indexes?

First of all storing such a volume of webpages, I would consider NOSQL(MongoDB) database.
Then apply indexes on those webpages.

Now execute queries to parse a given phone number.

we can use db language to get tht by giving commands and we can grt that data

you can use bucket sorting