Load data using Kafka connector

StarRocks provides a self-developed connector named Apache Kafka® connector (StarRocks Connector for Apache Kafka®) that continuously consumes messages from Kafka and loads them into StarRocks. The Kafka connector guarantees at-least-once semantics.

The Kafka connector can seamlessly integrate with Kafka Connect, which allows StarRocks better integrated with the Kafka ecosystem. It is a wise choice if you want to load real-time data into StarRocks. Compared with Routine Load, it is recommended to use the Kafka connector in the following scenarios:

• The format of source data is, for example, Protobuf, not JSON, CSV, or Avro.
• Customize data transformation, such as Debezium-formatted CDC data.
• Load data from multiple Kafka topics.
• Load data from Confluent cloud.
• Need finer control over load batch sizes, parallelism, and other parameters to achieve a balance between load speed and resource utilization.

Preparations

Set up Kafka environment

Both self-managed Apache Kafka clusters and Confluent cloud are supported.

• For a self-managed Apache Kafka cluster, make sure that you deploy the Apache Kafka cluster and Kafka Connect cluster and create topics.
• For Confluent cloud, make sure that you have a Confluent account and create clusters and topics.

Install Kafka connector

Submit the Kafka connector into Kafka Connect:

• Self-managed Kafka cluster:

  • Download and unzip starrocks-kafka-connector-1.0.0.tar.gz.
  • Copy the extracted directory to the libs directory of Kafka. Restart Kafka Connect to read the latest JAR files.

• Confluent cloud:

  NOTE

  The Kafka connector is not currently uploaded to Confluent Hub. You need to upload the compressed file to Confluent cloud.

Create StarRocks table

Create a table or tables in StarRocks according to Kafka Topics and data.

Examples

The following steps take a self-managed Kafka cluster as an example to demonstrate how to configure the Kafka connector and start the Kafka connect in order to load data into StarRocks.

1. Create a Kafka connector configuration file named connect-StarRocks-sink.properties and configure the parameters. For detailed information about parameters, see Parameters.

   name=starrocks-kafka-connector
   connector.class=com.starrocks.connector.kafka.StarRocksSinkConnector
   topics=dbserver1.inventory.customers
   starrocks.http.url=192.168.xxx.xxx:8030,192.168.xxx.xxx:8030
   starrocks.username=root
   starrocks.password=123456
   starrocks.database.name=inventory
   key.converter=io.confluent.connect.json.JsonSchemaConverter
   value.converter=io.confluent.connect.json.JsonSchemaConverter

   NOTICE

   If the source data is CDC data, such as data in Debezium format, and the StarRocks table is a Primary Key table, you also need to configure transform in order to synchronize the source data changes to the Primary Key table.

2. Run the Kafka Connector. For parameters and description in the following command, see Kafka Documentation.

   • Standalone mode

     bin/connect-standalone worker.properties connect-StarRocks-sink.properties [connector2.properties connector3.properties ...]

   • Distributed mode

     NOTE

     It is recommended to use the distributed mode in the production environment.

   • Start the worker.

     bin/connect-distributed worker.properties

   • Note that in distributed mode the connector configurations are not passed on the command line. Instead, use the REST API described below to configure the Kafka connector and run the Kafka connect.

     curl -i http://127.0.0.1:8083/connectors -H "Content-Type: application/json" -X POST -d '{
     "name":"starrocks-kafka-connector",
     "config":{
         "connector.class":"com.starrocks.connector.kafka.SinkConnector",
         "topics":"dbserver1.inventory.customers",
         "starrocks.http.url":"192.168.xxx.xxx:8030,192.168.xxx.xxx:8030",
         "starrocks.user":"root",
         "starrocks.password":"123456",
         "starrocks.database.name":"inventory",
         "key.converter":"io.confluent.connect.json.JsonSchemaConverter",
         "value.converter":"io.confluent.connect.json.JsonSchemaConverter"
     }
     }

3. Query data in the StarRocks table.

Parameters

Limits

• It is not supported to flatten a single message from a Kafka topic into multiple data rows and load into StarRocks.
• The Kafka Connector's Sink guarantees at-least-once semantics.

Best practices

Load Debezium-formatted CDC data

If the Kafka data is in Debezium CDC format and the StarRocks table is a Primary Key table, you also need to configure the transforms parameter and other related parameters.

transforms=addfield,unwrap
transforms.addfield.type=com.starrocks.connector.kafka.transforms.AddOpFieldForDebeziumRecord
transforms.unwrap.type=io.debezium.transforms.ExtractNewRecordState
transforms.unwrap.drop.tombstones=true
transforms.unwrap.delete.handling.mode

In the above configurations, we specify transforms=addfield,unwrap.

• The addfield transform is used to add the __op field to each record of Debezium CDC-formatted data to support the StarRocks primary key model table. If the StarRocks table is not a Primary Key table, you do not need to specify the addfield transform. The addfield transform class is com.Starrocks.Kafka.Transforms.AddOpFieldForDebeziumRecord. It is included in the Kafka connector JAR file, so you do not need to manually install it.
• The unwrap transform is provided by Debezium and is used to unwrap Debezium's complex data structure based on the operation type. For more information, see New Record State Extraction.