[geomesa-users] GeoMesa 2.3.0 released

[Jim Hughes <jhughes@xxxxxxxx>]

Hi all,

The GeoMesa team here at CCRi is happy to announce the release of 
version 2.3.0. Artifacts are available from Maven Central and on 
GitHub[1]. For detailed information on key changes, refer to the release 
notes and upgrade guide[2]. LocationTech GeoMesa 2.3.0 works with 
GeoTools 20.x and GeoServer 2.14.x.

New in GeoMesa 2.3.0 is a Redis datastore implementation, which supports 
all the existing GeoMesa index types. It is designed for high frequency 
streaming data, as an alternative to the Kafka datastore for some use 
cases.

Another highlight in 2.3.0 is our improved Confluent integration, with 
support for Kafka topics managed by Confluent using Avro schemas and 
Kafka 2.1.1. We are continuing to expand GeoMesa’s integration with 
Confluent technologies like the Avro Schema Registry.

The GeoMesa FileSystem datastore has received several improvements 
around how it handles metadata, including an option to persist metadata 
to a relational database. This should provide a good deal of lift for 
certain use cases. More improvements are planned, as we continue working 
to provide a great strategy for data at rest.

Additionally, our core indexing logic has been refactored to be more 
flexible. We now support full user customization of the attributes used 
in each index, as well as multiple spatial and spatio-temporal indices 
on different attributes. Additionally, adding new index strategies has 
been greatly simplified, and we hope to eventually provide alternate 
implementations to support different use-cases, for example based on 
Google’s S2 library.

For GeoMesa’s HBase support, a number of improvements have come out of 
our use of massive HBase clusters on AWS EMR, backed by S3 storage.

For ingest, the GeoMesa NiFi processors now support NiFi expression 
language and variables. The GeoMesa converter library can now create 
composite converters for XML and JSON formats, as well as interact with 
Avro Schema registries. Additionally, the command-line tools now support 
tar and zip archives directly.

With this release complete, the GeoMesa team is looking at some major 
upgrades for GeoMesa 3.0. As a project on the JVM, we have to work 
through the necessary steps to support Java 9 and later. Given that many 
of our datastores support and interact with Hadoop, we are looking at 
upgrades to Hadoop 3.0.x. Along with that, we are planning upgrades for 
Scala 2.12, Spark, HBase 2.x, and Accumulo 2.0.

As always, feel free to be in touch with us on our mailing lists and 
Gitter.

Cheers,

Jim

1. Binary downloads: 
https://github.com/locationtech/geomesa/releases/tag/geomesa_2.11-2.3.0
2. Release notes: 
https://geomesa.atlassian.net/wiki/spaces/GEOMESA/pages/605356033/GeoMesa+2.3.0+Release+Notes
Upgrade guide: http://www.geomesa.org/documentation/user/upgrade.html