How To Accelerate Data Warehouse Migrations With SnowConvert AI

Snowflake for Developers GuidesHow To Accelerate Data Warehouse Migrations With SnowConvert AI

How To Accelerate Data Warehouse Migrations With SnowConvert AI

Chanin Nantasenamat, Liam Sosinsky, Federico Zoufaly

Overview

At its core, data migration is the process of moving data between locations like on-premises servers, cloud platforms, or SaaS applications. This is typically driven by key business events like modernizing legacy infrastructure, centralizing data for analytics, or preparing datasets for AI training. However, an unsuccessful migration can lead to data loss, business disruptions and compliance issues.

SnowConvert AI and Snowpark Migration Accelerator are both tools from Snowflake designed to assist in code migration, but they serve different purposes.

SnowConvert AI is an AI-powered solution for migrating entire data ecosystems, including code and objects, from a wide range of legacy data warehouses and BI tools (such as Teradata, Oracle, and Google BigQuery) to Snowflake. Its core function is to intelligently automate the conversion of SQL-based workloads, reducing manual effort. This guide provides a complete, end-to-end walkthrough of a SnowConvert AI migration project, from initial setup and extraction to code conversion, deployment, and final data validation.

In contrast, the Snowpark Migration Accelerator is a specialized, free tool for migrations from Databricks or another Spark platform. Briefly, it converts existing Apache Spark codebases (in Python or Scala) to leverage the Snowpark API.

What You'll Learn

This guide will walk you through the complete end-to-end migration process, which is broken down into four key parts:

Part 1: Preparation and Setup

Setup and Installation
Understanding SnowConvert AI Projects

Part 2: Code Extraction and Conversion

Database Object Extraction
Code Conversion and Assessment
AI Verification
SSIS Replatform
Power BI Repointing

Part 3: Migration Execution

Deployment to Snowflake
Data Migration
Data Validation

Part 4: Post-Migration and Resources

Troubleshooting
Best Practices
Conclusion and Resources
Next Steps

What You'll Build

By the end of this quickstart guide, you will have a complete understanding of the SnowConvert AI migration workflow and will be able to execute end-to-end database migrations from legacy platforms to Snowflake, including data extraction, code conversion, deployment, data migration, and validation.

Prerequisites

Valid Snowflake account with appropriate permissions
- Ensure your target role has the appropriate privileges. Specifically, the role needs:
  1. The CREATE MIGRATION privilege at the account level. You can grant this with the following command:
    GRANT CREATE MIGRATION ON ACCOUNT TO ROLE <your_role>;
  2. The ability to create databases and objects within them.
Access to source database system (Teradata, Oracle, SQL Server, etc.)
- To prepare for the migration, verify that the role for your legacy data source has the necessary permissions. The role must have:
  1. Read access to all objects you plan to migrate.
  2. The ability to extract the Data Definition Language (DDL) code for these objects.
Company domain email address for access code generation
Windows 11+, macOS 13.3+, or Linux operating system
Minimum 4GB RAM
Internet connectivity for tool updates and licensing
Multi-factor authentication (MFA) enabled for Snowflake account

Setup and Installation

System Requirements

Before installing SnowConvert AI, ensure your system meets the minimum requirements:

Windows:

Windows 11 or newer
Minimum 4GB RAM (8GB+ recommended)

macOS:

macOS Ventura 13.3.1 or newer
Minimum 4GB RAM

Linux:

Any modern Linux distribution
Minimum 4GB RAM

Obtaining SnowConvert AI

SnowConvert AI can be downloaded from this link. It's highly recommended to complete the free course "SnowConvert AI for Conversion" available at Snowflake Training, which provides both overview and hands-on technical training.

IMPORTANT:

SnowConvert AI is free for all users

Download Snowconvert AI from any Snowflake account via Snowsight by navigating to Ingestion/Migrations in the left-hand menu.

Professional services are available for additional migration support. Learn more and contact a representative here.

Installing SnowConvert AI

Windows Installation

Download the .exe installer from the Snowflake portal
Double-click the downloaded file to start the installation
Follow the installation wizard prompts
Launch SnowConvert AI from the Applications menu

macOS Installation

Download the .dmg file from the Snowflake portal
Double-click the .dmg file to mount it
Drag SnowConvert AI to the Applications folder
Launch the application from Applications

Understanding Access Codes

To ensure quality, access codes are required to use SnowConvert AI and are valid for 3 months. Starting with version 1.2.0, the tool automatically generates access codes and supports all source platforms with a single code. For more information on access codes, visit the SnowConvert AI - Download and Access page.

Obtaining an Access Code

Method 1: Through the Application

Open SnowConvert AI
Click "Get an Access Code" link on the Project Creation page

Note: If you face connectivity issues due to restrictive firewall settings, please refer to this document for troubleshooting guidance

Fill out the required information form
Submit and wait for email confirmation

Method 2: Help Menu

Launch SnowConvert AI
Go to main menu bar
Click "Get an access code" option
Complete the form and submit

Understanding SnowConvert AI Projects

The process begins when a user creates a new project and assigns it a name. The user then selects the source code to be migrated; in the provided example, the source is SQL Server. To utilize SnowConvert, a free access code is required, which can be requested and is sent immediately via email. A local working folder must also be set up on the user's machine for the project. In addition to code conversion, the tool offers optional features to replatform ETLs, like Informatica or SSIS, and to repoint Power BI reports to the new Snowflake database.

What is a SnowConvert Project?

A SnowConvert AI Project (.snowct file) is a local file that stores:

Source platform configuration
Conversion settings
Project status and progress
Assessment and conversion results

Each time you "Save & Start Assessment", SnowConvert AI creates a .snowct project file. The file is saved in the same folder as your input source code and it contains all project configurations and status.

Here are some of the benefits of SnowConvert AI project files:

Resume work from where you left off
Maintain consistent conversion settings
Track progress across migration phases
Share project configurations with team members

Project Setup

To initiate the migration process:

Launch SnowConvert AI
Click "Create New Project"
Select source platform (Teradata, Oracle, SQL Server, etc.)
Choose input folder containing source code
Select output folder for converted code
Enter access code

When resuming the migration process:

Double-click on the .snowct file. This will open SnowConvert AI.
Use "Open Project" button in the application
Access recent projects via File → Open Recents

Database Object Extraction

Extraction Overview

After the project setup, the extraction phase begins. For a source like SQL Server, SnowConvert AI can connect directly to the database catalog to extract the Data Definition Language (DDL). As an alternative, code can be accessed from a local drive or a Git repository. During the extraction, SnowConvert AI connects to the source database and pulls the definitions for objects such as tables, views, and stored procedures. Once complete, it provides a summary of all the objects that were successfully extracted, such as the number of databases, schemas, and tables.

SnowConvert AI provides direct database connectivity to extract objects from source systems, currently supporting SQL Server and Amazon Redshift.

For other sources, you will still need to extract your SQL code into .sql files. You can use database export scripts here: https://github.com/Snowflake-Labs/SC.DDLExportScripts

SQL Server Extraction

SnowConvert AI offers flexible authentication options to seamlessly integrate with your existing SQL Server environments. The platform supports two primary methods, ensuring secure and efficient access to your databases:

Standard SQL Server Authentication: Uses a username and password.
Windows Authentication (Integrated Security): Leverages the Windows user's security credentials, often preferred in enterprise environments for seamless, secure integration with existing Windows security policies.

Security Options:

Trust Server Certificate (required if database uses trusted certificates)
Encrypt Connection (required for encrypted connections)

Extractable Objects:

Tables (structure and metadata)
Views (definitions and dependencies)
Functions (scalar and table-valued)
Stored procedures (complete logic)

Amazon Redshift Extraction

Authentication Methods:

IAM Provisioned Cluster
IAM Serverless
Standard authentication

Extractable Objects:

Tables
Views
Materialized views
Stored procedures

Extraction Process

Configure Connection:

Select authentication method
Provide connection credentials
Configure security settings if required

Connect to Database:

Click "Connect to Database"
Wait for "Connection established" notification
Browse available databases and schemas

Select Objects:

Expand database and schema trees
Select specific objects for extraction
Review object dependencies

Execute Extraction:

Click "Extract Objects"
Monitor extraction progress
Review extraction results

Validate Results:

Click "View Last Extraction Results"
Confirm all required objects extracted successfully
Note any extraction failures or warnings

-- Example extracted object structure
-- Original SQL Server procedure
CREATE PROCEDURE GetCustomerOrders
    @CustomerID INT
AS
BEGIN
    SELECT OrderID, OrderDate, TotalAmount
    FROM Orders
    WHERE CustomerID = @CustomerID
END

-- Will be prepared for conversion to Snowflake format

Code Conversion and Assessment

Conversion Process Overview

SnowConvert AI performs both assessment and conversion in a single unified process, generating comprehensive reports and converted code simultaneously.

Step 1: Conversion Settings

Encoding: Select appropriate character encoding (UTF-8 default)
Custom Schema: Specify target schema name if different from source
Database Name: Set custom database name for converted objects
Target Language: Choose SnowScript or JavaScript for procedures
Comments: Enable commenting of nodes with missing dependencies

Step 2: Execute Conversion

Click "Save & Start Assessment"
Monitor conversion progress
Review completion status
Examine generated reports

Understanding Conversion Results

Code Completeness Metric:

Score below 100% indicates missing object references
Review dependencies before deployment
Address missing references in source code

Conversion Status Indicators:

Green: Successfully converted, ready for deployment
Yellow: Functional Data Model (FDM) warnings present
Red: Equivalent Work Item (EWI) errors require attention

Assessment Reports:

Conversion summary statistics
Object-by-object conversion status
Complexity analysis and recommendations
Migration effort estimates

-- Example conversion result
-- Original Teradata SQL
SELECT 
    customer_id,
    SUM(order_amount) as total_orders
FROM orders
WHERE order_date >= ADD_MONTHS(CURRENT_DATE, -12)
GROUP BY customer_id;

-- Converted Snowflake SQL
SELECT 
    customer_id,
    SUM(order_amount) as total_orders
FROM orders
WHERE order_date >= DATEADD(MONTH, -12, CURRENT_DATE)
GROUP BY customer_id;

AI Verification

Overview

The AI Verification step (currently in Public Preview) adds a layer of intelligence to migrations by using AI agents to automatically verify and resolve errors that occur after the deterministic conversion step (the Code Conversion step is executed by a transpiler). The AI fixes are grounded with tests over synthetically generated data. This improves accuracy, reduces manual effort, and makes migrations faster, more reliable, and more scalable. AI will attempt to execute the converted code and fix issues if it is not working.

Step-by-Step flow

After the code conversion step customers can opt to use AI Verification to improve the quality of the converted code. If a customer is not allowed to use AI, this step can be skipped and continue to the Deployment of the converted objects ( as you can see in the image below, SKIP AI VERIFICATION button shows in the lower left corner of the screen). To continue with AI Verification we need to select which objects we want to verify. Sometimes certain objects will be auto selected by SnowConvert AI because they are dependencies of the selected objects. So don't be scared if you select one procedure but the tool auto selects a few tables, views and even other procedures. This means all those objects are required to be able to execute the selected stored procedure.

Once objects are selected, the VERIFY CODE can be clicked, which will display a special screen with a set of disclaimers. It is important to read and understand these disclaimers before proceeding. The AI will be executed in the customer Snowflake account using a Cortex Complete function. The tool will use the same Snowflake connection that was used for Source Code Extraction and that will be used for Deployment and Data Migration.

Once the user accepts all of the disclaimers, they can proceed with the actual verification. This step may take a long time as AI needs to upload files to a Snowflake stage to complete the verification process. At the end of the process, you will get a result screen similar to the image below.

At the top of your screen, you'll find a summary of the AI Results, followed by detailed information for each validated object. To gain a deeper understanding of how the AI reviewed a specific object, click "SEE DETAILS" for that object. This will display an explanation similar to the example shown in the following image.

It is important to understand that the migration engineer needs to review the AI results and merge them with the initial code conversion results.

SSIS Replatform

Overview

SSIS Replatform is a specialized SnowConvert AI capability designed to convert SQL Server Integration Services (SSIS) .dtsx packages to Snowflake-native solutions. This feature transforms complex ETL workflows into modern data pipelines using Snowflake tasks, procedures, and dbt projects.

For optimal conversion results, ensure that your input includes valid .dtsx files and all dependent database objects (DDL scripts for tables, views, functions, and procedures referenced by the SSIS packages).

The replatform process separates SSIS packages into two components:

Control Flow Orchestration: Converted to SQL scripts with Snowflake tasks and stored procedures.
Data Flow Tasks: Transformed into individual dbt projects for each data flow component.

Project Setup for SSIS Replatform

Launch SnowConvert AI
Click "Create New Project"
Important: Select "SQL Server" as the source platform
In the extraction configuration section, select "Replatform" option
Enter the path to your SSIS .dtsx files

DTSX Conversion Process

Proceed with the conversion process.
SnowConvert AI analyzes all .dtsx files and dependencies
Conversion proceeds following the standard SnowConvert AI workflow
Control flow tasks converted to Snowflake stored procedures and tasks
Data flow tasks converted to individual dbt projects

The converted ETL code will be available in the SnowConvert AI output folder under the ETL directory. An etl_configuration folder will also be created containing utilities and configuration files. Additionally, ETL.Elements.NA.csv and ETL.Issues.NA.csv files provide detailed information about the ETL conversion process and any issues encountered.

Output Structure

output/
└── ETL/
    └── [same folder structure as input]/
        └── [Package_Name]/
            ├── Orchestration.sql
            ├── Data Flow Task 1/
            │   ├── dbt_project.yml
            │   ├── models/
            │   └── sources/
            └── [Additional Data Flow Tasks]/

Understanding Conversion Results

Review the generated CSV files for detailed conversion information:

ETL.Elements.NA.csv: Details about converted ETL elements
ETL.Issues.NA.csv: Issues encountered during conversion

Deployment Process

Deploy Snowflake tasks and stored procedures
Deploy the dbt projects in your environment
Manually fix the failing components and validate model compilation and execution

Power BI Repointing

Overview

Power BI repointing is a SnowConvert AI feature that allows users to redefine connections from different database sources to a Snowflake valid connector. Currently, the supported sources are:

It is recommended that DDLs should be migrated with the Power BI repointing migration process, which would allow SnowConvert AI to identify objects such as tables and views.

Power BI Repointing Set Up

Save your Power BI projects with .pbit extension.
Open SnowConvert AI.
Optionally, add your DDLs.
Select the source language that you are using in your Power BI reports, i.e. SQL Server.
Add your .pbit files on the Power BI repointing section.
Click on the “continue to conversion” button.

Power BI Repointing Results

Open your repointed Power BI report.
Fill the Power BI parameters with your Snowflake account information: server link, warehouse name, and database name. (This is automatically added by SnowConvert AI).
Refresh your data in your reports.
Double-check results.
Save your repointed report with the format you need. i.e. .pbix.

Assessment

To review the overall status of the repointed reports, you can review the “ETLAndBiRepointing” assessment report. This report contains the report name and a summary of the changed connectors.

Deployment to Snowflake

Deployment Overview

SnowConvert AI facilitates the direct deployment of converted code objects and schemas to your Snowflake environment. This process includes comprehensive dependency management and precise execution ordering, ensuring a smooth and accurate transition.

Authentication Methods

SSO Authentication:

Integrates with the organization's Single Sign-On (SSO) provider for a seamless authentication experience.
Supports enterprise identity management systems.

Standard Authentication:

Username and password authentication
Requires multi-factor authentication (MFA) enabled
Account identifier format: orgname-account-name

Pre-Deployment Validation

Step 1: Review Conversion Status

Examine status indicators for each object (e.g., successful, needs review).
Review Converted Code: Examine the generated Snowflake SQL for key objects (e.g., procedures, functions) to confirm the logic is correctly translated before deployment.
Resolve any EWI (Error with Impact) errors before deployment.
Review FDM (Further Development Mandatory) warnings and decide if deployment is acceptable.
Use "Refresh Tree" to update status after code changes

MANDATORY: Resolve any EWI (Error with Impact) errors before deployment. These errors represent logic that could not be reliably converted and will cause deployment failure or runtime errors if not corrected.

Handling EWI Errors:
SnowConvert AI is designed to flag these high-impact errors, but cannot always fix them. To resolve an EWI, you must:

Examine the code snippet in the SnowConvert AI interface that triggered the EWI or better yet in your preferred Source Code IDE. Learn more

Manually modify the converted source code to fix the issue..

Unit test on the corrected file.

Object Selection:

Select only successfully converted objects

Consider deployment dependencies

Group related objects for batch deployment

Step 2: Initiate Deployment

Select objects for deployment
Click "Deploy" button
Choose authentication method
Provide Snowflake credentials

Step 3: Authentication

-- Example Snowflake connection details
Account: myorg-myaccount
Username: migration_user
Warehouse: MIGRATION_WH
Database: TARGET_DB
Schema: PUBLIC
Role: MIGRATION_ROLE

Step 4: Execution Order Deployment follows this automatic sequence:

Databases (create container structure)
Schemas (organize object namespaces)
Tables (establish data structures)
Views (depend on table structures)
Functions (provide reusable logic)
Stored Procedures (may reference other objects)

Step 5: Monitor Progress

Track deployment status for each object
Review execution logs and results
Address any deployment errors immediately

Deployment Results

Success Indicators:

Object created successfully in Snowflake
Proper dependency resolution
Correct object permissions applied

Error Handling:

Review detailed error messages
Fix source code issues if needed
Re-run conversion if necessary
Retry deployment for failed objects

-- Example deployed object in Snowflake
CREATE OR REPLACE PROCEDURE get_customer_orders(customer_id INTEGER)
RETURNS STRING
LANGUAGE SNOWSCRIPT
AS
$$
BEGIN
    LET result_cursor CURSOR FOR 
        SELECT order_id, order_date, total_amount
        FROM orders
        WHERE customer_id = :customer_id;
    
    RETURN 'Procedure executed successfully';
END;
$$;

Data Migration

Data Migration Overview

SnowConvert AI provides automated data migration capabilities to transfer actual data from Teradata, Oracle, SQL Server, Amazon Redshift, Synapse, Sybase, BigQuery, Netezza, Postgres, Greenplum and Databricks SQL source tables to deployed Snowflake structures, with platform-specific optimization strategies.

For migrations from Databricks or another Spark platform, you’ll use Snowpark Migration Accelerator.

Amazon Redshift Data Migration

Migration Strategy: Redshift data is migrated using S3 as an intermediary storage layer:

Data unloaded from Redshift to PARQUET files in S3
Files copied directly from S3 to Snowflake tables
Temporary files cleaned up automatically

Prerequisites:

S3 Bucket Configuration:

Bucket in same region as Redshift cluster
Empty bucket path (no existing files)
Proper IAM permissions configured

IAM Role for Redshift:

This JSON snippet defines an AWS Identity and Access Management (IAM) role specifically for Amazon Redshift, which is crucial for data migration, particularly when Redshift data is migrated using S3 as an intermediary storage layer.

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "s3:PutObject",
        "s3:GetObject",
        "s3:GetBucketLocation",
        "s3:ListBucket"
      ],
      "Resource": [
        "arn:aws:s3:::your-bucket-name/*",
        "arn:aws:s3:::your-bucket-name"
      ]
    }
  ]
}

IAM User for S3 Access:

This JSON snippet defines an AWS Identity and Access Management (IAM) user policy for S3 Access. This policy outlines the permissions granted to an IAM user to interact with a specific Amazon S3 bucket.

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "s3:PutObject",
        "s3:GetObject",
        "s3:GetObjectVersion",
        "s3:DeleteObject",
        "s3:DeleteObjectVersion",
        "s3:ListBucket",
        "s3:GetBucketLocation"
      ],
      "Resource": [
        "arn:aws:s3:::your-bucket-name/*",
        "arn:aws:s3:::your-bucket-name"
      ]
    }
  ]
}

Migration Process:

The migration process outlines the essential steps for efficiently transferring data from Redshift to Snowflake, ensuring data integrity and performance.

Configure S3 Settings:
- S3 Bucket URL (must end with "/")
- IAM Role ARN for data unloading
- Access Key ID and Secret Access Key
Select Tables:

Choose tables for data migration
Review table sizes and complexity
Consider migration batch sizing

Execute Migration:

Monitor unload progress from Redshift
Track S3 file creation and validation
Observe copy operations to Snowflake

Validate Results:

Compare row counts between source and target
Verify data integrity and completeness
Review migration timing and performance

SQL Server Data Migration

Migration Strategy: Direct data transfer using optimized connection protocols:

Bulk data extraction from SQL Server
Streaming transfer to Snowflake tables
Real-time progress monitoring

Prerequisites:

Valid SQL Server connection with read permissions
Network connectivity to both systems
Sufficient bandwidth for data transfer

Migration Process:

Verify Connections:

Test source SQL Server connectivity
Confirm Snowflake authentication
Validate table structures match

Select Tables:

Choose tables for migration
Review dependencies and foreign keys
Plan migration order if needed

Execute Transfer:

Start data migration process
Monitor transfer progress and speed
Handle any connection or timeout issues

Validate Migration:

Compare row counts automatically
Review data type conversions
Verify constraint compliance

Data Validation

Validation Overview

SnowConvert AI provides comprehensive data validation to ensure migration accuracy through schema validation and metrics validation, giving you confidence in your migrated data.

Validation Prerequisites

Python Requirements:

Python version ≥ 3.9 and < 3.13
Python available in system PATH
Virtual environment support

# Verify Python installation
python --version

# Expected output format: Python 3.x.x

Validation Modes

Schema Validations: Verifies structural integrity of migrated tables:

Table names match exactly
Column names preserved correctly
Ordinal positions maintained
Data types converted appropriately
Character lengths preserved
Numeric precision and scale maintained
Row counts match between source and target

Metrics Validations: Compares aggregate data metrics between source and target:

Minimum and maximum values
Average calculations
Null value counts
Distinct value counts
Standard deviation
Variance calculations
Custom business metrics

Validation Process

Prerequisites: Data validation requires Python ≥ 3.10 or < 3.13 to be installed and available in your PATH. To verify that Python is available in your PATH, run python --version in your terminal (or Command Prompt on Windows). Download Python here.

Step 1: Environment Setup

Data migration completion required
Python environment validation
Virtual environment creation
Dependency installation

Step 2: Validation Execution

Click "Validate Data" in the migration interface
System checks Python availability
Sets up isolated validation environment
Executes both schema and metrics validations

Step 3: Results Analysis Validation results are categorized into three levels:

Category	Description	Action Required
Pass	Values match exactly	No action needed
Warning	Minor differences (e.g., higher precision)	Reconcile data by: Applying transformation Changing the ingestion process
Fail	Values don't match	Investigation required

Validation Reports

Summary Table:

High-level validation status per table
Quick identification of issues
Overall migration success metrics

Detailed CSV Reports:

Column-by-column validation results
Specific discrepancy details
Sharable with stakeholders
Suitable for audit trails

Report Structure:

table_name,column_name,validation_type,source_value,target_value,status,notes
customers,customer_id,row_count,10000,10000,PASS,Counts match
orders,order_total,avg,156.78,156.78,PASS,Averages match
products,price,max,999.99,999.99,PASS,Maximum values match

Troubleshooting

Common Installation Issues

Permission Problems:

macOS .config issues: Ensure .config is a directory, not a file
Windows AppData access: Verify Snowflake Inc folder permissions
Linux installation: May require sudo for system-wide installation

Network Connectivity:

Firewall blocking updates: Whitelist sctoolsartifacts.blob.core.windows.net
Access code generation: Ensure access to https://snowconvert.snowflake.com/licensing
License validation: Check connectivity to licensing APIs

Version Management:

SnowConvert AI updates automatically when you launch it with an internet connection, but you can manually check for updates through the application's menu. If an update is available, the system will prompt you to download and install it. To verify the update is complete, you can check again in the menu to see the new version.
If facing any issues during the update, please uninstall previous versions and relaunch.

Performance Optimization

Large Dataset Migration:

Plan migrations during off-peak hours
Use incremental migration strategies for very large tables
Monitor system resources during migration
Consider table partitioning for optimal performance

Resource Management:

Ensure adequate memory (8GB+ recommended for large projects)
Monitor disk space for temporary files
Use SSD storage for better I/O performance
Close unnecessary applications during migration

Security Considerations

Access Control:

Use principle of least privilege for database connections
Enable MFA on all Snowflake accounts
Regularly rotate access codes and credentials
Audit migration activities and access logs

Data Protection:

Encrypt sensitive data during transit
Use secure connection protocols (SSL/TLS)
Implement proper backup strategies
Maintain audit trails for compliance

-- Example security configuration
-- Grant minimal required permissions
GRANT USAGE ON WAREHOUSE migration_wh TO ROLE migration_role;
GRANT CREATE TABLE ON SCHEMA target_db.public TO ROLE migration_role;
GRANT INSERT, SELECT ON ALL TABLES IN SCHEMA target_db.public TO ROLE migration_role;

Monitoring and Alerting

Migration Monitoring:

Track conversion success rates by object type
Monitor data migration progress and timing
Set up alerts for migration failures
Document migration performance baselines

Post-Migration Monitoring:

Validate application connectivity to new environment
Monitor query performance in Snowflake
Track user adoption and satisfaction
Plan for ongoing maintenance and optimization

IMPORTANT:

Always test conversions in development environments first

Maintain detailed migration documentation for future reference

Plan rollback strategies for critical production migrations

Engage Snowflake Professional Services for complex migrations

Best Practices

Adhering to best practices is crucial for a successful and efficient migration to Snowflake. This section outlines key considerations across SSIS replatforming, general migration strategies, data validation, and code conversion, providing guidance to optimize performance, ensure data integrity, and streamline the overall process.

SSIS Replatform

Document SSIS package dependencies and custom components.
Review ETL.Elements.NA.csv and ETL.Issues.NA.csv for conversion details.

Migration

Performance Optimization:

Migrate large tables during off-peak hours
Use parallel migration for multiple small tables
Monitor network bandwidth utilization
Consider table partitioning for very large datasets

Error Handling:

Implement retry logic for transient failures
Log detailed error information for troubleshooting
Maintain rollback capabilities where possible
Validate partial migrations before continuing

Data Integrity Checks:

-- Example validation query
-- Compare row counts between source and target
SELECT 
    'Source Count' as source,
    COUNT(*) as row_count 
FROM source_database.schema.table_name
UNION ALL
SELECT 
    'Target Count' as source,
    COUNT(*) as row_count 
FROM target_database.schema.table_name;

NOTE: Should you require additional help resolving errors and ensuring data integrity, please email [email protected]

Validation

Pre-Validation Checklist:

Ensure data stability during validation (no updates)
Complete all migration steps before validation
Have sufficient system resources available
Plan validation during maintenance windows

Interpreting Results:

Focus on FAIL status items first
Investigate WARNING items for business impact
Document any acceptable discrepancies
Rerun validation after fixes

Common Issues and Solutions:

Issue	Cause	Solution
Row count mismatch	Incomplete migration	Re-run data migration
Precision differences	Data type conversion	Verify acceptable business impact
Date format variations	Timezone or format changes	Standardize date handling
Null handling differences	Platform-specific null behavior	Update conversion rules

# Example validation check (conceptual)
def validate_table_metrics(source_table, target_table):
    validations = {
        'row_count': compare_row_counts(source_table, target_table),
        'column_stats': compare_column_statistics(source_table, target_table),
        'data_types': validate_schema_compatibility(source_table, target_table)
    }
    return validations

Conversion

Code Preparation:

Clean up source code before conversion
Remove commented-out legacy code
Ensure consistent encoding across all files
Document complex business logic

Project Organization:

Use meaningful project names and descriptions
Organize source code in logical folder structures
Maintain backup copies of original source code
Document custom conversion settings

Quality Assurance:

Review conversion reports thoroughly
Test converted code in development environment
Address EWI errors before deployment
Validate complex business logic manually

Conclusion And Resources

This guide has walked you through the complete lifecycle of data warehouse migration using SnowConvert AI, from initial setup and configuration to advanced topics like SSIS replatforming and AI verification. You are now equipped to confidently accelerate your migration projects, ensuring accuracy and efficiency as you transition to Snowflake's powerful platform.

What You Learned

Installation and Setup: Successfully installed SnowConvert AI across different operating systems
Access Management: Obtained and managed access codes for tool activation and learned troubleshooting techniques for licensing issues
End-to-End Migration: Executed complete database migrations including extraction, conversion, deployment, data migration, and validation
Best Practices: Applied security, performance, and quality assurance practices for successful enterprise-grade migrations
Troubleshooting: Identified and resolved common issues related to installation, conversion, and deployment processes

Related Resources

Articles:

SnowConvert AI for Conversion Training Course - Comprehensive hands-on training
What is Data Migration? Types, Strategy & Best Practices - Learn what data migration is, including types, strategies, best practices and data migration solutions for smooth data transfer and management.
Best Practices for Migrating Historical Data to Snowflake - Best practices for extracting, transferring, uploading, and validating data when migrating to Snowflake.
9 Best Practices for Transitioning From On-Premises to Cloud with Snowflake - This article provides nine best practices for a smooth and efficient transition from on-premises data systems to Snowflake's cloud data platform.

Documentation:

SnowConvert AI Documentation: Complete technical reference
SnowConvert AI Release Notes: Recent release notes

Next Steps

Now that you have successfully migrated your data warehouse to Snowflake using SnowConvert AI, consider these next steps to fully leverage the power of the Snowflake Data Cloud:

Explore Advanced Snowflake Features: Dive deeper into Snowflake's capabilities, such as advanced analytics with Snowpark, data sharing with other organizations, and building data applications directly on Snowflake.
Integrate with BI Tools: Connect your newly migrated Snowflake data to your existing or new Business Intelligence (BI) tools (e.g., Power BI, Tableau, Looker) to create insightful dashboards and reports.
Optimize Performance and Cost: Continuously monitor your Snowflake usage to optimize performance and costs.
Implement Data Governance and Security: Establish robust data governance policies and enhance security measures within Snowflake using features like row-access policies, column-level security, and data masking.
Automate Data Pipelines: Transition to automated data ingestion and transformation pipelines using Snowflake features like Openflow and Snowpipe.
Leverage AI/ML Capabilities: Integrate Snowflake with other cloud services and tools for machine learning and data science.
Application development: Native Apps and Streamlit Apps for building interactive data apps.
Stay Updated with Snowflake Innovations: Regularly check Snowflake's release notes, documentation, and community forums to stay informed about new features and best practices.

Updated 2026-03-20

This content is provided as is, and is not maintained on an ongoing basis. It may be out of date with current Snowflake instances