What's New RavenDB 7.0

RavenDB 7.0 introduces vector search and indexing, enabling advanced semantic and AI-driven queries by storing and comparing vector embeddings. This addition expands RavenDB’s indexing capabilities to cover use cases like semantic search, recommendation engines, and personalized experiences. By analyzing embeddings in vector form, RavenDB can capture subtle relationships between documents or entities, delivering more nuanced search results and enhanced data insights. Meanwhile, its robust architecture ensures high-performance retrieval and scalability for both small and large datasets.

With RavenDB, you can now integrate AI embedding generation effortlessly into your document workflows. Connect to any external embedding provider, such as OpenAI, Azure OpenAI, Hugging Face, Google AI, Ollama, or custom models, and generate vector embeddings automatically as part of your ETL or data processing pipeline. These embeddings are stored directly within your documents and can be queried using RavenDB’s built-in vector search, eliminating the need for extra infrastructure or sync layers. This approach simplifies deploying semantic search, recommendation systems, and AI-driven features across cloud, on-premises, and edge environments while keeping your data architecture unified and production-ready.

RavenDB Snowflake ETL provides direct data transfers from RavenDB to Snowflake: specify a connection string and JS transform script to start the ETL process immediately. It’s ideal for distributed setups, enabling local RavenDB processing of operational data and sending only relevant information to Snowflake for large-scale analytics. Common scenarios include real-time insights at the edge, efficient IoT data handling with centralized analysis, and extracting analytics from operational data to leverage Snowflake’s performance and cost benefits.

RavenDB 7.0 expands its ETL features by integrating AWS SQS, a cloud messaging service for communication between application components. This new capability enables seamless interaction with AWS cloud infrastructure. You can define JS scripts to transform documents and enqueue them directly into SQS, simplifying the implementation of reliable messaging patterns in distributed systems. RavenDB’s robust architecture ensures dependable asynchronous delivery and processing.

In Studio, alongside “All Documents,” there is now an “All Revisions” view that shows every revision across all documents or those within a specific collection. You can filter revisions by Collection or Type (Regular or Deleted, the latter created when a document is deleted). Any listed revision can be permanently deleted, and from this view, you can also review their content. This unified overview simplifies inspecting and managing revisions in your database.