I have similar problem about the internal indexing data structure According to Paolo Ferragina of Univ Pisa, B+tree with cluster is best for sorting. However, referring to the implementation of org.apache.lucene.search.IndexSearch, it looks like the impl doesn't take B+tree, never mention cluster.
Can anyone explain this issue in more depth? thanks, Kan --- shailesh kumar <[EMAIL PROTECTED]> wrote: > I had looked at the document you had listed as > well as used a Hex editor to look at the segment > files. .That is how I came to know about the > lexicographic sorting. But was not sure if BTree is > used. If I understand correctly a Binary tree (i.e > each node only 2 children) or a high order > Balanced tree (where in a range of values are stored > in the node and each node can have more than 2 > children) is the best way to search. > So wantted to know if Lucene implements it that > way ( if not in the data storage of the index in > the file, atleast in the memory during lookups??) > > Shailesh > > > > > > > --------------------------------- > Yahoo! Photos > Ring in the New Year with Photo Calendars. Add > photos, events, holidays, whatever. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
