Packt Publishing (http://www.packtpub.com/) have recently published a
book on SOLR by David Smiley and Eric Pugh. This is quite exciting as
up until now there have not been any books on SOLR, whereas Lucene
users have the excellent Lucene In Action. This book has been of
interest to DigitalPebble, as SOLR is one of the technologies that
DigitalPebble provides consultancy on and uses for its projects. As
you probably know, SOLR (htpp://lucene.apache.org/solr) is an
Entreprise Search Server based on Lucene and is an extremely
successful Apache project. SOLR adds a large number of functionality
to Lucene and, unless you have a very specific use case (or a tendency
for reinventing the wheel), you are more likely to prefer it to Lucene
if you need to expose the search functionality as a web application.

The book is based on version 1.4 of SOLR which has not yet been
released officially, however it is still a useful resource even if you
decide to stick to version 1.3. One of the main attractions of the
book is that it takes a very practical look at SOLR and uses examples
based on a real-world dataset. Some of the main differences between a
document based search system like SOLR and relational databases are
identified and commented: this will certainly be useful for users with
a background in the latter. Note that no preliminary knowledge of
Lucene or SOLR is necessary.

The book provides a good coverage of all major aspects of SOLR and its
use and should help most beginners with setting up, configuring and
fine-tuning their SOLR application. It contains many useful tips on
various aspects such as security, score debugging, integration with
third-party library or scalability. The proof-reading could have been
a bit more thorough, but the occasional spelling mistake or slight
inaccuracy does not prevent it from being very readable and
informative.