- Più recenti
- Maggior numero di voti
- Maggior numero di commenti
Yes, you can build a semantic search solution for your RDS MySQL database using Amazon OpenSearch Service. There are several approaches to connect RDS MySQL with OpenSearch:
-
AWS Database Migration Service (DMS): This is a recommended approach for continuously loading data from RDS to OpenSearch. DMS allows you to set up both full load and continuous data capture (CDC) from your RDS MySQL database directly to OpenSearch. This is ideal for keeping your search index synchronized with your database.
-
Data processing pipeline options:
- RDS → DMS → OpenSearch (direct approach)
- RDS → DMS → S3 → Lambda → OpenSearch (for custom transformations)
- RDS → DMS → S3 → OpenSearch Ingestion Service → OpenSearch (managed approach)
The second and third options are useful when you need to transform your data before indexing it in OpenSearch, such as concatenating columns, removing PII information, or other modifications.
Amazon OpenSearch Ingestion Service is particularly valuable as it's a fully managed, serverless data collector that can filter, transform, and enrich your records before publishing them to OpenSearch. It eliminates the need to manage your own pipeline infrastructure, automatically scales based on your defined capacity limits, and provides security patches and updates.
Once your data is in OpenSearch, you can leverage its vector database capabilities for semantic search, using k-nearest neighbor (k-NN) search with vector embeddings. OpenSearch supports SQL queries through its SQL plugin, allowing you to query your data using familiar SQL syntax via the /plugins/_sql endpoint. This can be helpful if you're transitioning from a relational database mindset.
For visualization and analysis, you can use OpenSearch Dashboards (included with the service) or connect through JDBC to business intelligence tools like Amazon QuickSight or SQL clients such as DBeaver.
Sources
Amazon RDS with Amazon Opensearch Service | AWS re:Post
Community | Search and Analyze data in Amazon OpenSearch using SQL
Introducing Point in Time queries and SQL/PPL support in Amazon OpenSearch Serverless | AWS Big Data Blog
Use DeepSeek with Amazon OpenSearch Service vector database and Amazon SageMaker | AWS Big Data Blog
I have a transactional database built in AWS RDS MySQL that deals with a typical orders data and their workflow management. I want to build semantic search capabilities on my Web application using the above data wherein the user wants to search not just on the columns but also the data within the tables using semantic search. Suggest if the above approach you suggested works? I also have another data source that comes to my postgres db on redshift via Kinesis firehose. How do I use it in opensearch?