Over on Apache.org they announced that Apache Lucene/Solr™ 4.1 is now available.

Lucene/Solr Version 4.1 supports significant new features, Admin UI improvements, storage improvements, enhancements to Solr Cloud, and includes numerous optimizations and bugfixes. Kudos to the large Lucene/Solr community! Everyone’s input and testing makes for a very robust product release.

Here’s a quick overview of some of the best improvements/enhancements:
Admin User Interface
* Improved Data Import Handler UI
* XML formatting easier to read by default
* Improved Core management
* Better IE support

Sharding Improvements
* Short circuiting for distributed search (only hits nodes that will have answers)
* Custom Sharding
* CompositeId router, domain/prefix routing)
* New API calls to control collections and shards, other admin API enhancements.

Technical Advancements
* Improved highlight indexing via storeOffsetsWithPositions
* Solr 4 geospatial now works with {!geofilt} and {!bbox} parsers
* Near-realtime support was added to the facet module. * New highlighter – PostingsHighlighter support
* Distributed search support for “find similar documents” (MLT / MoreLikeThis)
* Improved performance statistics, including percentile request times in various increments
* Improved advanced query parsing via localParams; no longer needs _query_ magic field
* Default UTF-8 support under various web containers
* Dynamic core loading and unloading
* Faster searches for rare values, such as Primary Key

What do you think are the best new features with Lucene/Solr 4.1?