I've added term cardinalities and result sizes.

-Eric

On Tue, Jan 31, 2012 at 3:28 PM, Keith Turner <[email protected]> wrote:

> For all of the words used in example queries, I would like to know how
> many documents each of those words occurred.  For example how many
> documents did 'old' occur in?
>
> The write up lets us know how many documents a query matched.  Does
> the query time include retrieving the document?  Or does it just
> return the document id?  If the document is returned, it would be nice
> to know the data size.  For example ('old' and 'man' and 'sea')
> returned 22,956 documents totaling 4.2M in size.
>
> Keith
>
> On Mon, Jan 30, 2012 at 3:26 PM, Eric Newton <[email protected]>
> wrote:
> > Hi Accumulators,
> >
> > In preparation for 1.4, we have included a more sophisticated example
> that
> > uses many of the unique Accumulo features.  I wrote up a little
> > introduction to the document-partitioned indexing design pattern, and
> > posted some performance numbers on a small test cluster.
> >
> > The example provides a simple web-based application that performs
> > un-weighted, multi-word searches over the contents of wikipedia.
> >
> > http://incubator.apache.org/accumulo/example/wikisearch.html
> >
> > Feedback is welcome.
> >
> > -Eric
>

Reply via email to