Keyrus enablement of Alteryx on AWS includes services to help your organization abide by any of the following best practices.

1. DEVELOPMENT SETUP & DATA CATALOGUE Create an organized directory that will be shared across logical groups of developers. Leverage generic macros that house common business logic to remove the risk of data quality problems. Work with a versioning software like GitLab or VisualStudio to keep a history of development and to better control deployments. Use Alteryx Connect for global search capabilities across apps,data sources, and external front-end dashboards (Tableau, Qlik, Quicksight) as well as Data Lineage and Data Dictionary.

2. SOFTWARE DEVELOPMENT LIFECYCLE Design: Gather business requirements and make sure to build the app to incorporate scale and flexibility. Reuse previous work when possible. Development: Develop Alteryx apps and workflows according to best practices. Integration/Regression Testing (if applicable): Build processes to automatically compare expected output against actual output of apps. Post-Development Activities: QA, code review, and UAT are essential to a successful rollout of an app.

3. DOCUMENTATION Annotation: Each tool in the workflow should be clearly annotated so the progression of data from A to Z makes sense. Containers/Comment Boxes: Logically group sections of the workflows to provide a description of what is happening to the data within said groups. These can also be used against input/output tools to quickly see what sources the workflow is reading and writing to. Workflow Description: Provide a workflow description in a comment box in the top-right portion of the workflow Create an automated process to parse workflow xml and produce flat file documentation of all workflows.

4. DEPLOYMENT Automate deployment through Alteryx REST API calls. Admin will be responsible for tagging workflows that are ready for deployment. If the Alteryx ecosystem is not using Gallery, you can leverage versioning platforms to automate the pull of production code to the locked-down folder. Build further customization into deployment to parse xml of workflows being deployed and check for certain rules (i.e., no packaged user data should be moving to Gallery).

5. Server Setup Multi-Node environment to handle large concurrency and resource-intensive workflows. High Availability: Create metadata repository replica set and make that the persistence layer for Alteryx. Have a “cold” stand-by controller machine to activate if the active controller goes down. Disaster Recovery: Take daily backups of metadata repository and place in a persistent storage.