21.5. Using the GeoMesa Accumulo Spark Runtime¶
geomesa-accumulo-spark-runtime module (found in the
geomesa-accumulo directory in the GeoMesa source distribution) provides a shaded JAR with all of the dependencies for Spark and Spark SQL analysis for data stored in GeoMesa Accumulo, including the GeoMesa Accumulo data store, the GeoMesa Spark and GeoMesa Spark SQL modules, and the
In the example below,
$VERSION = 1.3.2.
The shaded JAR can either be built from source (see Building from Source), or is included in the
dist/spark directory of the GeoMesa Accumulo binary distribution. This JAR may be passed as an argument to the Spark command line tools, or to Jupyter running the Toree kernel, without having to also include the other dependencies requred to run GeoMesa or Accumulo.
For example, the JAR can be used with the
--jars switch of the
$ spark-shell --jars /path/to/geomesa-accumulo-spark-runtime_2.11-$VERSION.jar
You should see a logging message loading the shaded JAR as the Spark console comes up:
... 17/01/16 14:31:21 INFO SparkContext: Added JAR file:/path/to/geomesa-accumulo-spark-runtime_2.11-$VERSION.jar at http://192.168.3.14:40775/jars/geomesa-accumulo-spark-runtime_2.11-$VERSION.jar with timestamp 1484595081362 ...