Home› Getting Started› Quick Start Guide› Deploy

Deploy

After synchronization, the data warehouse structure exists in the target database. The next step is deployment, where AnalyticsCreator generates and distributes deployment artifacts to the selected environment.

Deployment packages the generated database objects together with orchestration components such as pipelines and analytical models. This allows the data warehouse to be executed and used in a target environment such as SQL Server, Azure or Fabric.

Purpose

Package and deploy generated database structures, pipelines, and analytical models to a target environment.

Design Principle

Deployment separates structure generation from environment distribution.

  • Synchronization creates the structure
  • Deployment distributes and activates it in a target system

All deployment artifacts are generated from metadata and can be recreated at any time.

Inputs / Outputs

Inputs

  • Synchronized data warehouse model
  • Deployment configuration (target server, database, credentials)
  • Selected components (database objects, pipelines, semantic models)

Outputs

  • Deployment package containing:
    • SQL scripts or DACPAC
    • SSIS packages or Azure Data Factory pipelines
    • Analytical models (e.g. tabular model for Power BI)
  • Deployed artifacts in the target environment

Internal Mechanics

1. Deployment package creation

AnalyticsCreator generates a deployment package that contains all required components for the data warehouse. This includes database objects, pipeline definitions, and optional analytical models.

2. Target configuration

Deployment settings define where the artifacts will be deployed. This includes:

  • SQL Server or Azure environment
  • Database name
  • Authentication details

3. Database deployment

The generated database structure is applied to the target system. This may include:

  • Creating or updating schemas
  • Deploying tables, views, and procedures

4. Pipeline generation

AnalyticsCreator automatically generates orchestration components:

  • SSIS packages for on-premise environments
  • Azure Data Factory pipelines for cloud environments
  • Fabric Data Factory pipelines 

These pipelines define how data is extracted, transformed, and loaded.

5. Analytical model generation

If configured, a semantic model is generated and deployed. This includes:

  • Dimensions and measures
  • Relationships between tables
  • Compatibility with reporting tools such as Power BI

6. Deployment logging

The deployment process produces logs that show which objects and components were created or updated.

Types / Variants

Deployment targets

  • On-premise SQL Server
  • Azure SQL Database
  • Azure Synapse or Fabric environments

Pipeline variants

  • SSIS packages
  • Azure Data Factory pipelines

Analytical outputs

  • Tabular models for Power BI
  • Power BI project
  • Other supported analytical engines

Example

A deployment is configured with:

  • Target SQL Server database
  • SSIS package generation enabled
  • Tabular model generation enabled

After deployment:

  • Database objects are created in the target database
  • SSIS packages are generated and available in a Visual Studio project
  • A tabular model is deployed and available for Power BI

At this stage, the system is fully deployed but not yet populated with data.

When to Use / When NOT to Use

Use when

  • The model is finalized and synchronized
  • You want to move the data warehouse to a target environment
  • Pipelines and analytical models need to be generated

Do NOT assume deployment loads data

  • Deployment creates structure and pipelines
  • Data loading requires execution of pipelines

Performance & Design Considerations

  • Deployment time depends on model size and number of objects
  • Pipeline generation adds orchestration complexity but reduces manual work
  • Repeated deployments should be controlled via versioning

Design trade-off:

  • Automated deployment vs manual control of environment-specific configurations

Integration with other AnalyticsCreator features

  • Synchronization: provides the generated structure
  • Workflows: define execution order within pipelines
  • CI/CD: deployment packages can be integrated into pipelines
  • Repository: remains the source for regeneration

Common Pitfalls

  • Deploying without validating the model
  • Incorrect connection configuration
  • Assuming deployment includes data loading
  • Not selecting required pipeline or model components

Key Takeaway

Deployment packages and distributes the generated data warehouse structure, pipelines, and analytical models to a target environment, but does not execute data loading.