Migrating data from Apache Airflow to Databricks using CData SSIS Components.
Databricks is a unified data analytics platform that allows organizations to easily process, analyze, and visualize large amounts of data. It combines data engineering, data science, and machine learning capabilities in a single platform, making it easier for teams to collaborate and derive insights from their data.
The CData SSIS Components enhance SQL Server Integration Services by enabling users to easily import and export data from various sources and destinations.
In this article, we explore the data type mapping considerations when exporting to Databricks and walk through how to migrate Apache Airflow data to Databricks using the CData SSIS Components for Apache Airflow and Databricks.
Data Type Mapping
| Databricks Schema | CData Schema |
|---|---|
|
int, integer, int32 |
int |
|
smallint, short, int16 |
smallint |
|
double, float, real |
float |
|
date |
date |
|
datetime, timestamp |
datetime |
|
time, timespan |
time |
|
string, varchar |
If length > 4000: nvarchar(max), Otherwise: nvarchar(length) |
|
long, int64, bigint |
bigint |
|
boolean, bool |
tinyint |
|
decimal, numeric |
decimal |
|
uuid |
nvarchar(length) |
|
binary, varbinary, longvarbinary |
binary(1000) or varbinary(max) after SQL Server 2000 |
Special Considerations
- String/VARCHAR: String columns from Databricks can map to different data types depending on the length of the column. If the column length exceeds 4000, then the column is mapped to nvarchar (max). Otherwise, the column is mapped to nvarchar (length).
- DECIMAL Databricks supports DECIMAL types up to 38 digits of precision, but any source column beyond that can cause load errors.
Prerequisites
- Visual Studio 2022
- SQL Server Integration Services Projects extension for Visual Studio 2022
- CData SSIS Components for Databricks
- CData SSIS Components for Apache Airflow
Create the project and add components
-
Open Visual Studio and create a new Integration Services Project.
- Add a new Data Flow Task to the Control Flow screen and open the Data Flow Task.
-
Add a CData Apache Airflow Source control and a CData Databricks Destination control to the data flow task.
Configure the Apache Airflow source
Follow the steps below to specify properties required to connect to Apache Airflow.
-
Double-click the CData Apache Airflow Source to open the source component editor and add a new connection.
-
In the CData Apache Airflow Connection Manager, configure the connection properties, then test and save the connection.
Start by setting the Profile connection property to the location of the ApacheAirflow Profile on disk (e.g. C:\profiles\ApacheAirflow.apip). Next, set the ProfileSettings connection property to the connection string for ApacheAirflow (see below).
ApacheAirflow API Profile Settings
Apache Airflow 3 uses JWT Bearer tokens for API authentication. You can generate a token from the Airflow web UI under Settings or via the Airflow CLI using airflow users create and the /api/v2/auth/token endpoint. Note that this profile targets Apache Airflow 3.x using the /api/v2 REST API. The legacy /api/v1 endpoint used by Airflow 2.x is not supported.
After setting the following connection properties, you are ready to connect:
- AuthScheme: Set this to APIKey.
- APIKey: Set this to your Apache Airflow JWT Bearer token.
- Server: Set this to the base URL of your Airflow instance (e.g. http://localhost:8080).
-
After saving the connection, select "Table or view" and select the table or view to export into Databricks, then close the CData Apache Airflow Source Editor.
Configure the Databricks destination
With the Apache Airflow Source configured, we can configure the Databricks connection and map the columns.
-
Double-click the CData Databricks Destination to open the destination component editor and add a new connection.
-
In the CData Databricks Connection Manager, configure the connection properties, then test and save the connection. 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, 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).
Other helpful connection properties
- QueryPassthrough: When this is set to True, queries are passed through directly to Databricks.
- ConvertDateTimetoGMT: When this is set to True, the components will convert date-time values to GMT, instead of the local time of the machine.
- UseUploadApi: Setting this property to true will improve performance if there is a large amount of data in a Bulk INSERT operation.
- UseCloudFetch: This option specifies whether to use CloudFetch to improve query efficiency when the table contains over one million entries.
-
After saving the connection, select a table in the Use a Table menu and in the Action menu, select Insert.
-
On the Column Mappings tab, configure the mappings from the input columns to the destination columns.
Run the project
You can now run the project. After the SSIS Task has finished executing, data from your SQL table will be exported to the chosen table.