SurrealDB Scalability
Blog post from SurrealDB
SurrealDB is a versatile, multi-paradigm database designed for performing document, graph, temporal, spatial, and text operations within an ACID-compliant environment, utilizing RocksDB as its storage layer for optimal write and compute performance. The database architecture separates storage and compute layers, allowing scalable solutions that align with the "pay for what you use" model, particularly beneficial in cloud environments. SurrealDB's scalability is enhanced by integrating TiKV, a scalable storage solution that distributes data across nodes, with Placement Drivers managing data location. The compute layer's statelessness enables easy scaling by adding or removing instances without complex coordination. Although currently lacking a query planner, SurrealDB supports complex IDs for predictable query performance and secondary tables to optimize data access patterns. It offers fault tolerance through a multi-raft consensus algorithm and a write-ahead log for data recovery, ensuring system reliability. While SurrealDB is ready for production use, ongoing user feedback, primarily from its Discord community, continues to shape its development and improve its features.