Stability and performance
With the release of SurrealDB 2.0, we're excited to introduce a brand-new parser for SurrealQL. This update offers significant improvements such as enhanced memory management, better support for recursive queries, and a more robust transaction layer with advanced caching capabilities. These upgrades ensure smoother query performance and improved efficiency for all your database operations.
Querying
With SurrealQL, you can leverage SurrealDB's multi-model capabilities, including querying graph edges, creating schemafull and schemaless definitions, performing vector embeddings with vector search, and executing full-text search. Additionally, we've introduced new features to further enhance the developer experience, as SurrealQL remains at the core of our product.
GraphQL
We're excited to announce that you can now query SurrealDB using GraphQL. With Surrealist, our intuitive user interface for SurrealDB, you can easily connect to any instance, run real-time queries, explore tables, and design schemas - all from one central platform.
Indexing
SurrealDB 2.0 introduces major indexing improvements, boosting performance with updated hashing and better write efficiency for BTree- and MTree-based indexes. The new ALTER statement with the REBUILD option, along with new CONCURRENT indexing functionality, optimises degraded indexes. Storing the HNSW algorithm on disk ensures fast, durable queries accessible through the DEFINE INDEX statement.
Security
One of the major updates in version 2.0 is a complete overhaul of SurrealDB's security features. Authentication is now more powerful and flexible than ever, including custom authentication hooks, cryptography settings and duration for tokens and sessions. Users now have more control over what the SurrealDB server is allowed to do and are able to better harden it against attackers. New security functions provide protection against HTML injection attacks and faster hashing algorithms.
SurrealKV
We're excited to debut SurrealKV in SurrealDB 2.0, our native key-value storage engine. Built entirely in Rust, like the rest of SurrealDB, SurrealKV is an embedded, ACID-compliant engine with built-in versioning. It supports historical and temporal querying, offering powerful, efficient data storage and retrieval.
SDKs
We're excited to announce significant improvements across all our SDKs, now in version 1.0 stable. These updates deliver enhanced performance, increased flexibility, and more robust support to meet your development needs, ensuring a smoother and more efficient experience for developers across the board.
SurrealML
SurrealML is a powerful engine that enables you to store and execute machine learning models directly within SurrealDB, right next to your data. You can integrate ML into your workflows without changing your tech stack, using models to compute calculations in queries. SurrealML also allows versioning, making it easy to compare model results and generate real-time insights.