Perform Batch Operations with Databricks Data in Apache NiFi

Apache NiFi supports powerful and scalable directed graphs of data routing, transformation, and system mediation logic. When paired with the CData JDBC Driver for Databricks, NiFi can work with live Databricks data. This article shows how to read data from a CSV file and perform batch operations (INSERT/UPDATE/DELETE) using the CData JDBC Driver for Databricks data in Apache NiFi (version 1.9.0 or later).

With built-in optimized data processing, the CData JDBC Driver offers unmatched performance for interacting with live Databricks data. When you issue complex SQL queries to Databricks, the driver pushes supported SQL operations, like filters and aggregations, directly to Databricks and utilizes the embedded SQL engine to process unsupported operations client-side (often SQL functions and JOIN operations). Its built-in dynamic metadata querying allows you to work with and analyze Databricks data using native data types.

Getting Started

---

Generate a JDBC URL

We need a JDBC URL to connect to Databricks data from Apachi NiFi.

Built-in Connection String Designer

For assistance in constructing the JDBC URL, use the connection string designer built into the Databricks JDBC Driver. Either double-click the JAR file or execute the jar file from the command-line.

Fill in the connection properties and copy the connection string to the clipboard.

To connect to a Databricks cluster, set the properties as described below.

Note: The needed values can be found in your Databricks instance by navigating to Clusters, and selecting the desired cluster, and selecting the JDBC/ODBC tab under Advanced Options.

• Server: Set to the Server Hostname of your Databricks cluster.
• HTTPPath: Set to the HTTP Path of your Databricks cluster.
• Token: Set to your personal access token (this value can be obtained by navigating to the User Settings page of your Databricks instance and selecting the Access Tokens tab).

Batch Operations (INSERT/UPDATE/DELETE) in Apache NiFi

The sample flow presented below is based on the following NiFi Processors:

• ListFile - Retrieves a listing of files from the local filesystem and creates a FlowFile for each retrieved file.
• FetchFile - Reads the content of the FlowFile received from the ListFile processor.
• PutDatabaseRecord - Uses a specified RecordReader to input records from a flow file coming from the FetchFile processor. These records are translated to SQL statements and executed as a single transaction.
• LogAttribute - Emits attributes of the FlowFile at the specified log level.

This is what our finished product looks like:

Disclaimers

1. The column names of the CSV file must match the column names of the data source table records to be inserted/updated/deleted.

2. Apache NiFi versions earlier than 1.9.0 do not support the Maximum Batch Size property in the PutDatabaseRecord processor.

Configurations

In order to perform batch INSERT, UPDATE or DELETE, the NiFi Processors should be configured similar to the following:

1. Configure the ListFile processor: Set the Input Directory property to the local folder path from where to pull the CSV files. Set the File Filter property to a regular expression to pick up only the files whose names match the expression. i.e., if the CSV file's full path is C:\Users\Public\Documents\InsertNiFi.csv, the properties should be configured like in the following image:

2. Configure the FetchFile processor
Leave the FetchFile processor's property configurations to their default values:
3. Configure the PutDatabaseRecord processor

INSERT operation

Configure the PutDatabaseRecord processor similar to the following in order to perform Batch INSERT Operations:


• Set the Record Reader property to a CSV Reader Controller Service. Configure the CSV Reader Controller Service to match the format of your CSV file.
• Set the Statement Type property to INSERT.
• Set the Database Connection Pooling Service to the DBCPConnection Pool that holds the driver configuration. Please note that the driver should be configured to use Bulk API.

  Property	Value
  Database Connection URL	jdbc:databricks:Server=127.0.0.1;Port=443;TransportMode=HTTP;HTTPPath=MyHTTPPath;UseSSL=True;User=MyUser;Password=MyPassword;
  Database Driver Class Name	cdata.jdbc.databricks.DatabricksDriver

• Set the Catalog Name property to the name of the catalog that your table is part of.
• Set the Schema Name property to the name of the schema that your table is part of.
• Set the Table Name property to the name of the table that you want to INSERT into.
• Set the Maximum Batch Size property to the maximum number of records that you want to be included in a single batch.

UPDATE Operation

Configure the PutDatabaseRecord processor similar to the following in order to perform Batch UPDATE Operations:

• Set the Record Reader property to a CSV Reader Controller Service. Configure the CSV Reader Controller Service to match the format of your CSV file.
• Set the Statement Type property to UPDATE.
• Set the Database Connection Pooling Service to the DBCPConnection Pool that holds the driver configuration. Please note that the driver should be configured to use Bulk API. Use the same Database Connection URL format as seen above.
• Set the Catalog Name property to the name of the catalog that your table is part of.
• Set the Schema Name property to the name of the schema that your table is part of.
• Set the Table Name property to the name of the table that you want to UPDATE.
• Set the Update Keys property to the name of the columns that are required for an UPDATE.
• Set the Maximum Batch Size property to the maximum number of records that you want to be included in a single batch.

DELETE Operation

Configure the PutDatabaseRecord processor similar to the following in order to perform Batch DELETE Operations:

• Set the Record Reader property to a CSV Reader Controller Service. Configure the CSV Reader Controller Service to match the format of your CSV file.
• Set the Statement Type property to DELETE.
• Set the Database Connection Pooling Service to the DBCPConnection Pool that holds the driver configuration. Please note that the driver should be configured to use Bulk API. Use the same Database Connection URL format as seen above.
• Set the Catalog Name property to the name of the catalog that your table is part of.
• Set the Schema Name property to the name of the schema that your table is part of.
• Set the Table Name property to the name of the table that you want to UPDATE.
• In comparison to INSERT and UPDATE statement types, the DELETE operation does not expose a Maximum Batch Size property. However, the operations are still processed in batches. If not changed, the maximum number of records per batch is 2000, adhering to the default value. In order to change the value of the Maximum Batch Size used for DELETE operations, change the statement type to INSERT or UPDATE, then change the value of the Maximum Batch Size property, and click Apply Changes. Finally, reopen the processor's configuration, change the Statement Type back to DELETE, and click Apply Changes.


4. Configure the LogAttribute processor

   Finally, configure the LogAttribute processor by specifying the Attributes that you would like to log or ignore, as well as the log level, based on your use case.

Free Trial & More Information

Download a free, 30-day trial of the CData JDBC Driver for Databricks and start working with your live Databricks data in Apache NiFi. Reach out to our Support Team if you have any questions.