GeoMesa 2.1.0 Release Notes

GeoMesa 2.1.0 is the first minor release in the 2.x line. This release includes 163 bug fixes and improvements since the 2.0.2 release. Users should read the 2.1 upgrade guide.

Dependency Version Updates

  • Apache Spark updated from 2.2 to 2.3 (2.2 is still supported)
  • Apache Arrow updated from 0.6 to 0.10
  • Typesafe Config updated from 1.2.1 to 1.3.3
  • json4s updated from 3.2.10 to 3.5.4
  • scalatra updated from 2.3.0 to 2.6.3
  • commons-compress updated from 1.4.1 to 1.16
  • Apache Orc updated from 1.4.1 to 1.4.4
  • specs2 updated from 2.6.3 to 4.3.2

Notable Features

  • Support for Spark 2.3 and improved Pyspark interoperability
  • Type inference for auto ingestion of data
  • Kafka 'event-time' message ordering and support for log compaction
  • Improved support for 'wide' schemas (ones with many attributes)
  • Apache Kudu data store implementation
  • FileSystem data store support for non-point geometries

Versions and Compatibility

The 2.x line of GeoMesa follows Semantic Versioning. It provides data compatibility with versions back to 1.2.0. It offers public API compatibility with versions 2.x. It offers binary compatibility with versions 2.1.x.

Note that the GeoMesa public API is not currently well defined. Users with deep code integration with GeoMesa may need to update some calls.

Upgrading

GeoMesa 2.1.0 supports upgrade in place from previous versions of 2.x, by updating the GeoMesa JAR versions. Ensure that JARs are updated in all clients and servers. See the upgrade guide for information on changes that may affect you after the update.