INTRODUCTION
Solbase is highly scalable search engine built on top of Lucene, Solr, and HBase. In a nut shell, Solbase replaced Lucene's index file layer that was stored in local file system under NFS to big database implementation in HBase.
Leveraging HBase's high performance scan, inverted index is efficiently stored in HBase for fast data retrieval.
http://www.slideshare.net/KyungseogOh/solbase
MAIN HIGHLIGHTS OF SOLBASE
Because HBase's ability to scale and distribute large data in distributed fashion, Solbase can scale and at the same time it perform very well with local in memory cache.
Solbase also inherited Solr's distributed processing layer for sharding large index into smaller pieces.
It also overcame Lucene & Solr's limitation with sortable fields. Instead of creating array of all docs for sorting/filtering, values are stored at index. therefore, Solbase can fetch those vals on the fly.
DEPENDENCIES:
By overcoming Lucene & Solr's limitations, Solbase had to forked off of Lucene and Solr repos for modifications and improvements in its cores. Solbase-Lucene & Solbase-Solr packages are not included in this repository, but can be found at:
https://github.com/Photobucket/Solbase-Solr https://github.com/Photobucket/Solbase-Lucene
PREREQUISITE:
Hadoop & HBase installed and running
Servlet container
INSTALLATION:
solr/home JNDI for schema.xml reference
populate Solbase tables in HBase
run ant script to generate solbase.war and put solbase.war in tomcat webapps dir, start tomcat
inserting schema information
RUNNING EXAMPLE:
you need to have Hadoop/HBase up and running as well as Tomcat with solbase.war
run bin/createTables.sh
run bin/insertSchemaInfo.sh books src/example/schema.xml
run runExampleImporter.sh on root dir of Solbase repo
http://localhost:8080/solbase/books/select?q=game&wt=xml&sort=price%20asc
NOTE: