Amazon Web Services (AWS) has introduced a significant enhancement to its AWS Database Migration Service Schema Conversion (AWS DMS SC), which promises to revolutionize the process of database schema conversion. This new feature is centered around the use of generative AI, a technology designed to streamline and automate the conversion of up to 90% of schema objects from commercial databases to PostgreSQL migrations. This development aims to simplify the migration process, making it more efficient and cost-effective.
Understanding AWS DMS and Schema Conversion
AWS Database Migration Service (AWS DMS) is a cloud-based service that facilitates the migration of databases from one system to another. It supports multiple types of databases, including relational databases, data warehouses, and NoSQL databases. With AWS DMS, businesses can seamlessly migrate their data to the Amazon Web Services (AWS) Cloud, or between various cloud and on-premises configurations.
To date, AWS DMS has been instrumental in migrating over one million databases. Its primary function is to aid in the transfer of data from one database system to another. When different database engines are involved, AWS DMS SC plays a crucial role in converting the source database schema and procedures to align with the target database system.
Enhancing Schema Conversion with Generative AI
Despite AWS DMS SC’s ability to automate many migration steps, challenges remain, particularly when complex database code elements require manual intervention. This is often the case with proprietary system functions or procedures and data type conversions, which do not have direct equivalents in PostgreSQL. Such complexities can prolong migration timelines and increase costs.
The newly introduced generative AI capability is designed to address these issues by automating some of the most labor-intensive schema conversion tasks. By utilizing large language models (LLMs) hosted on Amazon Bedrock, this capability enhances the existing conversion processes. It is particularly effective in converting code snippets in the source database that were previously unsupported by traditional rule-based techniques, including complex procedures and functions.
The integration of generative AI into AWS DMS SC reduces migration costs and accelerates project timelines. By automating more of the schema conversion process, businesses can focus on more strategic tasks, such as refining and optimizing applications after migration, rather than manually handling conversion gaps. Early adopters of these AI-powered features have already reported success, experiencing both cost savings and faster migrations.
How the New Capability Works
To illustrate the ease of using this new generative AI feature, let’s delve into the schema conversion process in AWS DMS SC. The process begins with a self-managed commercial database running on Amazon Elastic Compute Cloud (Amazon EC2). Through the AWS Management Console, users define the instance profile and data providers, setting up the replication instance network details, database engine endpoints, and securely stored database passwords. A migration project is then created, setting the stage for schema conversion.
Once the project is established, users can launch the schema conversion tool from the Schema conversion tab. The first-time launch takes a few minutes. AWS DMS SC’s generative AI capability is an opt-in feature, requiring users to enable it via the Settings tab. This option allows users to activate the Enable Generative AI feature for conversion.
Before diving into the conversion details, users can assess the overall migration complexity by selecting the schema they intend to migrate and choosing the Assess option. A summary of the assessment is typically available within minutes, providing detailed insights on the Action items tab. Users can export the results into a PDF report, which is accessible from an S3 bucket and shareable with colleagues.
Leveraging Generative AI for Schema Conversion
The summary screen highlights the percentage of Database storage objects and Database code objects convertible by the rule-based method. For example, it may show a 100% conversion capability for storage objects and 57% for code objects. By employing generative AI, these conversion rates can be significantly improved.
The PDF report provides an executive summary, statistics on the number of objects to be migrated, the feasibility of conversion using generative AI, and the migration complexity. This information is invaluable for understanding the scope and challenges of the migration.
With no blockers detected for migrating stored procedures, users can select a specific procedure, use the Actions menu on the source database, and choose Convert. Within minutes, the original procedure code appears on the left pane, while the proposed migrated version is displayed on the right.
The summary screen updates to reflect that 100% of the code is now convertible automatically. Users can edit and adjust the code as needed. Once satisfied with the proposed version, they can apply changes via the Actions menu on the target database side.
Compliance and Flexibility
This generative AI capability allows AWS DMS SC to automatically convert a substantial portion of schema objects from commercial databases to PostgreSQL. To ensure compliance with various policies, the feature is initially turned off, and users can enable it as needed. The system flexibly decides between traditional rule-based methods and generative AI, based on the complexity of the objects being converted. Organizations with strict policies against generative AI can continue using the rule-based approach, with manual adjustments required for unconverted or partially converted objects.
Availability and Pricing
The generative AI-enhanced schema conversion capability is now available in specific AWS Regions, including US East (Ohio, N. Virginia), US West (Oregon), and Europe (Frankfurt). AWS DMS Schema Conversion with generative AI offers a faster pathway for migration, facilitating a smoother transition to AWS.
To explore this new capability, visit the AWS DMS Schema Conversion documentation, which provides detailed instructions on how generative AI can simplify your next database migration. This enhancement represents a significant leap forward in database migration technology, paving the way for more efficient and cost-effective transitions. For more information, you can refer to the AWS official website.
In conclusion, the integration of generative AI into AWS Database Migration Service Schema Conversion marks a transformative step in database migration. By automating complex tasks and reducing manual intervention, this capability not only accelerates migration timelines but also cuts costs, making it an attractive solution for businesses looking to transition to AWS.
For more Information, Refer to this article.
