Install on Microsoft Fabric

This tutorial will guide you through the process of installing Sedona on Microsoft Fabric Synapse Data Engineering's Spark environment.

Step 1: Open Microsoft Fabric Synapse Data Engineering

Go to the Microsoft Fabric portal and choose the Data Engineering option.

Step 2: Create a Microsoft Fabric Data Engineering environment

On the left side, click My Workspace and then click + New to create a new Environment. Let's name it ApacheSedona.

Step 3: Select the Apache Spark version

In the Environment page, click the Home tab and select the appropriate version of Apache Spark. You will need this version to install the correct version of Apache Sedona.

Step 4: Install the Sedona Python package

In the Environment page, click the Public libraries tab and then type in apache-sedona. Please select the appropriate version of Apache Sedona. The source is PyPI.

Step 5: Set Spark properties

In the Environment page, click the Spark properties tab, then create the following 3 properties:

• spark.sql.extensions: org.apache.sedona.viz.sql.SedonaVizExtensions,org.apache.sedona.sql.SedonaSqlExtensions
• spark.serializer: org.apache.spark.serializer.KryoSerializer
• spark.kryo.registrator: org.apache.sedona.core.serde.SedonaKryoRegistrator

Step 6: Save and publish the environment

Click the Save button and then click the Publish button to save and publish the environment. This will create the environment with the Apache Sedona Python package installed. The publishing process will take about 10 minutes.

Step 7: Find the download links of Sedona jars

1. Learn the Sedona jars you need from our Sedona maven coordinate
2. Find the sedona-spark-shaded jar from Maven Central. Please pay attention to the Spark version and Scala version of the jars. If you select Spark 3.4 in the Fabric environment, you should download the Sedona jars with Spark 3.4 and Scala 2.12 and the jar name should be like sedona-spark-shaded-3.4_2.12-1.5.1.jar.
3. Find the geotools-wrapper jar from Maven Central. Please pay attention to the Sedona versions of the jar. If you select Sedona 1.5.1, you should download the geotools-wrapper jar with version 1.5.1 and the jar name should be like geotools-wrapper-1.5.1-28.2.jar.

The download links are like:

https://repo1.maven.org/maven2/org/apache/sedona/sedona-spark-shaded-3.4_2.12/1.5.1/sedona-spark-shaded-3.4_2.12-1.5.1.jar
https://repo1.maven.org/maven2/org/datasyslab/geotools-wrapper/1.5.1-28.2/geotools-wrapper-1.5.1-28.2.jar

Step 8: Start the notebook with the Sedona environment and install the jars

In the notebook page, select the ApacheSedona environment you created before.

In the notebook, you can install the jars by running the following code. Please replace the jars with the download links of the 2 jars from the previous step.

%%configure -f
{
    "jars": ["https://repo1.maven.org/maven2/org/datasyslab/geotools-wrapper/1.5.1-28.2/geotools-wrapper-1.5.1-28.2.jar", "https://repo1.maven.org/maven2/org/apache/sedona/sedona-spark-shaded-3.4_2.12/1.5.1/sedona-spark-shaded-3.4_2.12-1.5.1.jar"]
}

Step 9: Verify the installation

You can verify the installation by running the following code in the notebook.

from sedona.spark import *


sedona = SedonaContext.create(spark)


sedona.sql("SELECT ST_GeomFromEWKT('SRID=4269;POINT(40.7128 -74.0060)')").show()

If you see the output of the point, then the installation is successful.

Optional: manually upload Sedona jars to the Fabric environment LakeHouse storage

If your cluster has no internet access or you want to skip the slow on-the-fly download, you can manually upload the Sedona jars to the Fabric environment LakeHouse storage.

In the notebook page, choose the Explorer and click the LakeHouses option. If you don't have a LakeHouse, you can create one. Then choose Files and upload the 2 jars you downloaded in the previous step.

After the upload, you should be able to see the 2 jars in the LakeHouse storage. Then please copy the ABFS paths of the 2 jars. In this example, the paths are

abfss://9e9d4196-870a-4901-8fa5-e24841492ab8@onelake.dfs.fabric.microsoft.com/e15f3695-af7e-47de-979e-473c3caa9f5b/Files/sedona-spark-shaded-3.4_2.12-1.5.1.jar

abfss://9e9d4196-870a-4901-8fa5-e24841492ab8@onelake.dfs.fabric.microsoft.com/e15f3695-af7e-47de-979e-473c3caa9f5b/Files/geotools-wrapper-1.5.1-28.2.jar

If you use this option, the config files in your notebook should be

%%configure -f
{
    "conf": {
        "spark.jars": "abfss://XXX/Files/sedona-spark-shaded-3.4_2.12-1.5.1.jar,abfss://XXX/Files/geotools-wrapper-1.5.1-28.2.jar",
    }
}