Using UDAs Overview

Introduction

Integrating a User-Defined Analytics (UDA) into a package involves organizing essential files, defining metadata, and configuring settings for deployment within kdb Insights.

Creating a UDA

To create a UDA:

1. Define the purpose: Determine the specific problem or task that your UDA addresses.

2. Develop the UDA: Write the UDA, ensuring your code is modular, efficient, and well-documented.

3. Test the UDA: Validate the UDA with different datasets to ensure it performs as expected in various scenarios.

4. Refine and optimize: Use test results to refine and optimize the UDA for performance and accuracy.

Adding to a Package

To package your UDA for deployment:

1. Organize files: Arrange all necessary files, including the UDA script and dependencies, in a structured manner.

2. Create a metadata file: Develop a manifest that lists the UDA and related components, including version information and metadata.

3. Configure settings: Specify required configuration settings, such as environment variables, to ensure the UDA functions correctly after deployment.

4. Test the package: Test the package in a local environment to confirm all components work together as expected.

Deploying

Deploy your UDA by following these steps:

1. Validate the package: Ensure the UDA package is complete and all dependencies are correctly referenced.

2. Deploy to staging: Test the UDA in a staging environment to confirm it integrates well with other components.

3. Deploy to production: Once validated, deploy the UDA to the production environment.

4. Verify deployment: After deployment, perform checks to confirm the UDA is functioning correctly and monitor its performance.