Quickstart: Manage data with Azure Cosmos DB Spark 3 OLTP Connector for API for NoSQL


This tutorial is a quick start guide to show how to use Azure Cosmos DB Spark Connector to read from or write to Azure Cosmos DB. Azure Cosmos DB Spark Connector supports Spark 3.1.x and 3.2.x.

Throughout this quick tutorial, we rely on Azure Databricks Runtime 10.4 with Spark 3.2.1 and a Jupyter Notebook to show how to use the Azure Cosmos DB Spark Connector.

You can use any other Spark (for e.g., spark 3.1.1) offering as well, also you should be able to use any language supported by Spark (PySpark, Scala, Java, etc.), or any Spark interface you are familiar with (Jupyter Notebook, Livy, etc.).


  • An Azure account with an active subscription.

  • Azure Databricks runtime 10.4 with Spark 3.2.1

  • (Optional) SLF4J binding is used to associate a specific logging framework with SLF4J.

SLF4J is only needed if you plan to use logging, also download an SLF4J binding, which will link the SLF4J API with the logging implementation of your choice. See the SLF4J user manual for more information.

Install Azure Cosmos DB Spark Connector in your spark cluster using the latest version for Spark 3.2.x.

The getting started guide is based on PySpark/Scala and you can run the following code snippet in an Azure Databricks PySpark/Scala notebook.

Create databases and containers

First, set Azure Cosmos DB account credentials, and the Azure Cosmos DB Database name and container name.

cosmosEndpoint = "https://REPLACEME.documents.azure.com:443/"
cosmosMasterKey = "REPLACEME"
cosmosDatabaseName = "sampleDB"
cosmosContainerName = "sampleContainer"

cfg = {
  "spark.cosmos.accountEndpoint" : cosmosEndpoint,
  "spark.cosmos.accountKey" : cosmosMasterKey,
  "spark.cosmos.database" : cosmosDatabaseName,
  "spark.cosmos.container" : cosmosContainerName,

Next, you can use the new Catalog API to create an Azure Cosmos DB Database and Container through Spark.

# Configure Catalog Api to be used
spark.conf.set("spark.sql.catalog.cosmosCatalog", "com.azure.cosmos.spark.CosmosCatalog")
spark.conf.set("spark.sql.catalog.cosmosCatalog.spark.cosmos.accountEndpoint", cosmosEndpoint)
spark.conf.set("spark.sql.catalog.cosmosCatalog.spark.cosmos.accountKey", cosmosMasterKey)

# create an Azure Cosmos DB database using catalog api
spark.sql("CREATE DATABASE IF NOT EXISTS cosmosCatalog.{};".format(cosmosDatabaseName))

# create an Azure Cosmos DB container using catalog api
spark.sql("CREATE TABLE IF NOT EXISTS cosmosCatalog.{}.{} using cosmos.oltp TBLPROPERTIES(partitionKeyPath = '/id', manualThroughput = '1100')".format(cosmosDatabaseName, cosmosContainerName))

When creating containers with the Catalog API, you can set the throughput and partition key path for the container to be created.

For more information, see the full Catalog API documentation.

Ingest data

The name of the data source is cosmos.oltp, and the following example shows how you can write a memory dataframe consisting of two items to Azure Cosmos DB:

spark.createDataFrame((("cat-alive", "Schrodinger cat", 2, True), ("cat-dead", "Schrodinger cat", 2, False)))\
  .toDF("id","name","age","isAlive") \

Note that id is a mandatory field for Azure Cosmos DB.

For more information related to ingesting data, see the full write configuration documentation.

Query data

Using the same cosmos.oltp data source, we can query data and use filter to push down filters:

from pyspark.sql.functions import col

df = spark.read.format("cosmos.oltp").options(**cfg)\
 .option("spark.cosmos.read.inferSchema.enabled", "true")\

df.filter(col("isAlive") == True)\

For more information related to querying data, see the full query configuration documentation.

Partial document update using Patch

Using the same cosmos.oltp data source, we can do partial update in Azure Cosmos DB using Patch API:

cfgPatch = {"spark.cosmos.accountEndpoint": cosmosEndpoint,
          "spark.cosmos.accountKey": cosmosMasterKey,
          "spark.cosmos.database": cosmosDatabaseName,
          "spark.cosmos.container": cosmosContainerName,
          "spark.cosmos.write.strategy": "ItemPatch",
          "spark.cosmos.write.bulk.enabled": "false",
          "spark.cosmos.write.patch.defaultOperationType": "Set",
          "spark.cosmos.write.patch.columnConfigs": "[col(name).op(set)]"

id = "<document-id>"
query = "select * from cosmosCatalog.{}.{} where id = '{}';".format(
    cosmosDatabaseName, cosmosContainerName, id)

dfBeforePatch = spark.sql(query)
print("document before patch operation")

data = [{"id": id, "name": "Joel Brakus"}]
patchDf = spark.createDataFrame(data)


dfAfterPatch = spark.sql(query)
print("document after patch operation")

For more samples related to partial document update, see the GitHub code sample Patch Sample.

Schema inference

When querying data, the Spark Connector can infer the schema based on sampling existing items by setting spark.cosmos.read.inferSchema.enabled to true.

df = spark.read.format("cosmos.oltp").options(**cfg)\
 .option("spark.cosmos.read.inferSchema.enabled", "true")\

# Alternatively, you can pass the custom schema you want to be used to read the data:

customSchema = StructType([
      StructField("id", StringType()),
      StructField("name", StringType()),
      StructField("type", StringType()),
      StructField("age", IntegerType()),
      StructField("isAlive", BooleanType())

df = spark.read.schema(customSchema).format("cosmos.oltp").options(**cfg)\

# If no custom schema is specified and schema inference is disabled, then the resulting data will be returning the raw Json content of the items:

df = spark.read.format("cosmos.oltp").options(**cfg)\

For more information related to schema inference, see the full schema inference configuration documentation.

Configuration reference

The Azure Cosmos DB Spark 3 OLTP Connector for API for NoSQL has a complete configuration reference that provides additional and advanced settings writing and querying data, serialization, streaming using change feed, partitioning and throughput management and more. For a complete listing with details see our Spark Connector Configuration Reference on GitHub.

Migrate to Spark 3 Connector

If you are using our older Spark 2.4 Connector, you can find out how to migrate to the Spark 3 Connector here.

Next steps