I'm building a small web search engine for searching about 1 million web pages and I want to know What is the best way to build the inverted index ? using the DBMS or What …? from many different views like storage cost, performance, speed of indexing and query? and I don't want to use any open source project for that I want to make my own one!
Perhaps you might want to elaborate why you do not wish to use F/OSS tools like Lucene or Sphinx.