incubator-couchdb-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Paul C. Nendick" <paul.nend...@gmail.com>
Subject Re: Lucene & CouchDB
Date Thu, 25 Sep 2008 13:41:29 GMT
This is an interesting question and one I hope is answered more fully
by someone more in the know than I.

>From a distance, the differences I see are:

- Lucene attempts to bring structure and searching to an otherwise
unstructured set of documents stored using any variety of means.
CouchDB is rather on the other end - it presents an easy-to-search
document store of unstructured documents.

- The view creation features of CouchDB is the closest analogue to
what Lucene does.

- To its detriment, CouchDB cannot (yet) distribute that view creation
task across machines like Lucene can (using Hadoop:
http://skillsmatter.com/podcast/home/distributed-lucene-for-hadoop).

- CouchDB is more buzzword-compliant what with its REST and its JSON

- Solr brings much of the last point to Lucene (http://lucene.apache.org/solr/)

Comments?

/p


2008/9/24 Ayende Rahien <ayende@ayende.com>:
> I am trying to understand the association between CouchDB and Lucene. I
> understand that Lucene can be used as the full text search engine for
> CouchDB.
> My question is more in terms of understanding the difference between them.
> Lucene itself is a document storage with many of the same semantics as
> CouchDB.
> CouchDB has views and replication, which Lucene doesn't have.
>
> Can anyone clear the difference between them?
>

Mime
View raw message