Install Sedona Scala/Java

Before starting the Sedona journey, you need to make sure your Apache Spark cluster is ready.

There are two ways to use a Scala or Java library with Apache Spark. You can user either one to run Sedona.

Spark interactive Scala or SQL shell: easy to start, good for new learners to try simple functions
Self-contained Scala / Java project: a steep learning curve of package management, but good for large projects

Spark Scala shell¶

Download Sedona jar automatically¶

Have your Spark cluster ready.

Run Spark shell with --packages option. This command will automatically download Sedona jars from Maven Central.

./bin/spark-shell --packages MavenCoordinates

Please refer to Sedona Maven Central coordinates to select the corresponding Sedona packages for your Spark version.

Local mode: test Sedona without setting up a cluster

./bin/spark-shell --packages org.apache.sedona:sedona-spark-shaded-3.0_2.12:1.5.0,org.apache.sedona:sedona-viz-3.0_2.12:1.5.0,org.datasyslab:geotools-wrapper:1.5.0-28.2

Cluster mode: you need to specify Spark Master IP

./bin/spark-shell --master spark://localhost:7077 --packages org.apache.sedona:sedona-spark-shaded-3.0_2.12:1.5.0,org.apache.sedona:sedona-viz-3.0_2.12:1.5.0,org.datasyslab:geotools-wrapper:1.5.0-28.2

Download Sedona jar manually¶

Have your Spark cluster ready.
Download Sedona jars:
- Download the pre-compiled jars from Sedona Releases
- Download / Git clone Sedona source code and compile the code by yourself (see Compile Sedona)
Run Spark shell with --jars option.
```
./bin/spark-shell --jars /Path/To/SedonaJars.jar
```
If you are using Spark 3.0 to 3.3, please use jars with filenames containing 3.0, such as sedona-spark-shaded-3.0_2.12-1.5.0; If you are using Spark 3.4 or higher versions, please use jars with Spark major.minor versions in the filename, such as sedona-spark-shaded-3.4_2.12-1.5.0.
- Local mode: test Sedona without setting up a cluster
```
./bin/spark-shell --jars /path/to/sedona-spark-shaded-3.0_2.12-1.5.0.jar,/path/to/sedona-viz-3.0_2.12-1.5.0.jar,/path/to/geotools-wrapper-1.5.0-28.2.jar
```
- Cluster mode: you need to specify Spark Master IP
```
./bin/spark-shell --master spark://localhost:7077 --jars /path/to/sedona-spark-shaded-3.0_2.12-1.5.0.jar,/path/to/sedona-viz-3.0_2.12-1.5.0.jar,/path/to/geotools-wrapper-1.5.0-28.2.jar
```

Spark SQL shell¶

Please see Use Sedona in a pure SQL environment

Self-contained Spark projects¶

A self-contained project allows you to create multiple Scala / Java files and write complex logics in one place. To use Sedona in your self-contained Spark project, you just need to add Sedona as a dependency in your pom.xml or build.sbt.

To add Sedona as dependencies, please read Sedona Maven Central coordinates
Use Sedona Template project to start: Sedona Template Project
Compile your project using SBT. Make sure you obtain the fat jar which packages all dependencies.
Submit your compiled fat jar to Spark cluster. Make sure you are in the root folder of Spark distribution. Then run the following command:
```
./bin/spark-submit --master spark://YOUR-IP:7077 /Path/To/YourJar.jar
```

Note

The detailed explanation of spark-submit is available on Spark website.

Last update: November 13, 2023 05:10:18