On Sep 27, 2007, at 11:59 AM, Ben wrote:
You might be better off, performance
wise, doing this in a C full-text indexing system and wrapping an
interface to it.
I hadn't thought of that yet. Can you recommend any?
Elephant's use of BDB is designed for certain kind of operations and
access patterns. Implementing a full text index means you are adding
a data structure abstraction on top of another one that wasn't
designed for it. The resulting awkward interaction can have
significant performance penalties. A low level data structure
explicitly designed for large-scale, fast read access inverse
indexing can outperform an 'interpreted' one significantly.
lucene is good, solar is a web service packaging of it, and there is a
port for common lisp (montezuma), ruby (ferret) and perl (kinosearch).
B
_______________________________________________
elephant-devel site list
elephant-devel@common-lisp.net
http://common-lisp.net/mailman/listinfo/elephant-devel
_______________________________________________
elephant-devel site list
elephant-devel@common-lisp.net
http://common-lisp.net/mailman/listinfo/elephant-devel
