Microsoft |Azure Synapse

Azure Synapse Analytics is a unified analytics platform that combines data warehousing, big data processing (via Apache Spark), and log/timeseries analysis. It brings together SQL, Spark, and Data Explorer capabilities into a single workspace—helping organizations analyze data efficiently and at scale.

Prerequisites

Ensure the following requirements are met before you connect Azure Synapse as a data source:

  • ADOC access & permissions to add data sources.
  • An existing data plane in ADOC or create one for Azure Synapse Analytics ingestion. See the Data Plane Installation documentation for more information.
  • Azure Synapse workspace details, including:
    • Workspace name
    • SQL Pool configuration (e.g., Dedicated SQL Pool)
    • Authentication credentials (SQL user/password, Managed Identity, or Service Principal)

Add Azure Synapse Analytics as a Data Source

Step 1: Start Setup

  1. From the main menu, select Register > Add Data Source.
  2. Select Azure Synapse Analytics from the list.
  3. On the Data Source Details page:
  4. Enter a name and optional Description for the data source.
  5. Enable the Data Reliability toggle and select the data plane.
  6. Click Next.

Step 2: Add Connection Details

  1. Enter your Azure Synapse workspace name.
  2. Select a SQL Pool configuration method (currently, Dedicated SQL Pool Name is supported).
  3. Authentication options:
  4. Username & Password: Provide the SQL Pool name, username, and password.
  5. Managed Identity (optional): Use for secure Azure credential-based access. If enabled, assign a Managed Identity to access Synapse via Azure credentials.
  6. Service Principal (optional): Use for Azure AD-based authentication with Client ID, Client Secret, and Tenant ID.
  7. Secret Manager (optional): Use when storing credentials in a secure secret store. Provide workspace, SQL Pool details, username, and secret configuration details.
  8. Select the Data Plane Engine: Choose either Spark or Pushdown for better performance during profiling and checks.
  9. Click Test Connection. If successful, proceed. Otherwise, double-check your entries and network connectivity.
  10. Click Next to proceed to configure observability.

Step 3: Set Up Observability

  1. Set a scan schedule, including frequency and time zone, so ADOC can crawl Synapse for metadata and quality metrics.
  2. Enable Notifications for crawler success or failure alerts.
  3. Click Submit to finalize the setup.

Troubleshooting and FAQs

1. Connection Error After Test Connection

  • Issue: ADOC fails to connect.
  • Solution: Verify workspace and SQL Pool names, authentication credentials, firewall settings, and whether you're using the correct Data Plane Engine.

2. Performance Slowness in Profiling

  • Issue: Profiling or data checks take too long.
  • Solution: Switch to Pushdown engine for faster execution using Synapse’s native compute.

3. Freshness Metrics Not Visible

  • Issue: No freshness or data latency information shows up.
  • Solution: Use a Row Count Check as a proxy metric for freshness, or rely on external solutions like ADF (Azure Data Factory) SLAs.

What’s Next

  • Profile your Azure Synapse Analytics data source to begin applying Data Reliability policies.
  • Create and apply observability policies: Define Data Quality, Schema Drift, and Anomaly Detection checks. (No dashboards until policies are in place.)
  • Configure alerts: Proactively notify your team on quality or pipeline issues.
Type to search, ESC to discard
Type to search, ESC to discard
Type to search, ESC to discard