Twitter reworks search engine for better results

The index of the new search engine is twice as big as its predecessor.

Twitter has launched a new backend architecture for search, the microblogging site announced on its Engineering blog on Wednesday.

The new backend search architecture translates to a bigger index -- twice as long as the current one -- without making the search slower, the blog added.

The engineers' team said the site is now on a new, modern search architecture that is based on a "highly efficient" inverted index instead of a relational database.

"Since we love open source here at Twitter we chose Lucene, a search engine library written in Java, as a starting point," the blog added.

The microblogging site also announced that it handles over 12,000 queries per second, which translates to more than a billion queries per day.

The team said it had to rewrite parts of Lucene's data structures to suit the new requirements of the site.