Aurora DSQL is a distributed SQL database service available in a free preview, replacing Amazon RDS Aurora. It uses PostgreSQL code for SQL processing over a distributed transaction and storage engine, unlike any other databases available under the Amazon Aurora umbrella.
This service has a different runtime behavior, necessitating adjustments to align with its runtime behavior. It is compatible with the PostgreSQL wire protocol and dialect, requiring AWS Identity and Access Management (IAM) for authentication.
The article provides a visual tutorial on how to connect to the Aurora DSQL serverless cluster using a PostgreSQL client, which requires generating a short-lived token for admin authentication.
The article also highlights the notable database features, including IAM mapping, admin role creation, and short-lived generated password format, providing advantages in matters of security and authentication.
The only allowed SSL mode on Aurora DSQL is 'require.' The article details how to create environment variables for the connection with the help of the AWS CLI.
This article provides a comprehensive guide to connecting to Aurora DSQL databases using a PostgreSQL client while also summarizing the main sources of confusion that can be encountered on the way.
This article serves as the first in a series of articles that will test and provide insights into Aurora DSQL's architecture and trade-offs in reproducible tests on the preview, available documentation, and publicly announced roadmap details.
The upcoming series will also explore the many features and details of Aurora DSQL and highlight how it differs from YugabyteDB, the other distributed SQL database using PostgreSQL code.