Create a FlexConnect Data Source

FlexConnect makes it simple to connect data from any source, no matter the format or protocol, into the GoodData platform. Many companies have data spread across different systems, making it hard to support all combinations with standard connectors. FlexConnect solves this by allowing you to connect to almost any data source, including streaming data, structured and unstructured datasets, databases, third-party platforms, and machine learning models.

The time it takes to deliver value is critical for any analytics project. Long delays caused by complex ETL processes can stall projects before they launch. FlexConnect eliminates these obstacles by connecting to data quickly without compromising governance or security.

What FlexConnect Offers

• Universal Data Source Connectivity

  FlexConnect connects to all types of data sources, including cloud-based, on-premises, hybrid systems, SQL/NoSQL databases, APIs, and machine learning models. It provides seamless access to data without needing third-party tools.

• Self-Service Integration

  Teams can build and manage their own data connections without vendor assistance. This approach speeds up pipeline creation, reduces delays, and enhances Time to Value.

• Embedded Data Processing

  With FlexConnect, you can perform transformations, apply business logic, or call machine learning operations directly during runtime. This ensures that business rules and logic are consistently applied.

• Integration with GoodData

  After processing, FlexConnect integrates data into GoodData’s semantic model. This ensures high-quality analytics across dashboards, APIs, and SDKs.

How FlexConnect Can Be Used

• Integrating External APIs

  FlexConnect provides custom connectors for third-party APIs or internal systems. These connectors integrate data directly into GoodData’s semantic model, eliminating the need for slow, batch-based processes.

  If you are interested, we highlight this use-case in an article about Weather Forecast.

• Connecting NoSQL Databases

  FlexConnect simplifies the process of integrating NoSQL databases like MongoDB. It transforms unstructured data into a dimensional model for a user-friendly analytics experience.

  Our article about MongoDB connection can help you understand how this works in practice.

• Real-Time Machine Learning Integration

  FlexConnect brings real-time machine learning insights into your analytics workflows, allowing businesses to make faster, more accurate decisions.

  If you want to see how to connect a pickled model, read our article on machine learning.

• Centralized Data Governance (Unity Catalog)

  Organizations using Unity Catalog can benefit from FlexConnect’s seamless integration while maintaining strong governance, data discovery, and collaboration.

  Learn how to connect tables from Unity Catalog to your BI tools. This guide provides a ready-to-use approach to getting started quickly and efficiently.

• Real-Time Analytics with Streaming Data

  FlexConnect supports real-time data sources like Apache Kafka, enabling businesses to analyze and act on data streams instantly, such as stock feeds or social media trends.

  To put this more into perspective, we wrote an article about financial transactions in Kafka.

How FlexConnect Works

FlexConnect works similarly to table functions found in many database systems. A table function is a user-defined function that returns a table and behaves like a database view but with added flexibility, such as taking parameters.

Key Benefits:

• Simple Contract: The function is called with a set of parameters and returns table data.
• Dynamic Adaptation: Functions can adapt based on context, such as the user viewing the data or applied filters.

Getting Started with FlexConnect:

1. Build Your Server: Use our template to set up your server. The template handles the setup so you can focus on the logic.
2. Deploy: Deploy your server using any preferred method.
3. Add the FlexConnect Data Source: Once the FlexConnect data source is added, you can work with the developed functions like regular data sets. You can create models and interact with FlexConnect data just like any other data source. Query context information is passed to the functions, allowing them to provide different data based on user groups and applied filters.