TL;DR
Benchmark tests show SurrealDB 3.x outperforms Postgres, MongoDB, Neo4j, and Redis in various workloads with full durability enabled. SurrealDB’s recent major updates have significantly improved its performance, especially in CRUD operations and full-table scans.
Recent benchmark tests reveal that SurrealDB 3.x significantly outperforms Postgres, MongoDB, Neo4j, and Redis in various workloads, with full disk durability enabled, on identical hardware. The results demonstrate substantial improvements in CRUD throughput, query latency, and scan performance, marking a major milestone for SurrealDB’s development.
The benchmarks used a single hardware setup featuring an AMD Ryzen Threadripper 9970X CPU, 128 GiB DDR5 RAM, and NVMe storage, running Ubuntu 24.04. All databases were configured for optimal production performance, with fsync enabled to ensure full durability. The tests involved 128 clients executing 48 concurrent queries on datasets with 5 to 15 million rows, including mixed data types and complex queries.
SurrealDB 3.x showed dramatic performance gains over earlier versions, with a 164-fold increase in full-table scan speed and a 3.2-fold improvement in indexed lookup speed compared to SurrealDB 2.x. The latest version achieved 141,000 CRUD operations per second, up from 107,000 in version 2.x, and 78,000 in version 1.x. Latency reductions ranged from 27% to nearly 100%, depending on the workload.
When compared to other databases, SurrealDB demonstrated superior write throughput. It was approximately 1.5 times faster than Postgres and 5 to 7 times faster than MySQL on create, update, and delete operations. In read-heavy workloads, SurrealDB was marginally faster than Postgres but slightly slower than MongoDB on single-record writes, though it outperformed MongoDB on read operations and unindexed filter scans, being roughly 2.7 times faster than MongoDB in unindexed predicate filtering.
Why It Matters
These results are significant because they position SurrealDB as a competitive, durable, multi-model database capable of handling high-throughput workloads efficiently. Its performance improvements could influence enterprise adoption, especially in scenarios requiring full durability and multi-model capabilities, challenging traditional relational and document databases. The benchmarks also highlight the importance of configuration and tuning in achieving optimal performance, emphasizing SurrealDB’s readiness for production environments.
Timetec 1TB SSD NVMe PCIe Gen3x4 8Gb/s M.2 2280 3D NAND High Performance SLC Cache Read/Write Speed Up to 2,000/1,600 MB/s Internal Solid State Drive for PC Laptop and Desktop (1TB)
NVMe PCIe Gen3x4 3D NAND Breakthrough Read and Write Speeds: Sequential Read Speed up to 2,000MB/s and Write…
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Background
SurrealDB has undergone major internal rewrites across three releases, focusing on query parsing, storage, and indexing. Previously, benchmarks with fsync disabled showed promising but less realistic performance metrics. The latest tests with full durability provide a more accurate comparison for production use. SurrealDB’s development aims to bridge the gap with established databases like Postgres and MongoDB, offering a multi-model approach with transactional guarantees. The benchmarks follow a recent trend of rigorous, hardware-consistent testing to ensure fair comparisons across diverse database systems.
“The recent performance gains in SurrealDB 3.x reflect our commitment to building a robust, high-performance multi-model database suitable for production workloads.”
— SurrealDB team
“Running all databases on the same hardware with production-grade configuration and full durability enabled provides a fair comparison that reflects real-world performance.”
— Benchmark author
AMD Ryzen™ Threadripper™ 9970X
AMD Ryzen Threadripper Processors for Desktop Workstations
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What Remains Unclear
It is still unclear how SurrealDB’s performance scales with larger datasets, more complex queries, or different hardware configurations. The benchmarks focus on specific workloads and configurations, so real-world results may vary. Additionally, the impact of future software updates on performance remains to be seen.
Crucial 32GB DDR5 RAM Kit (2x16GB), 5600MHz (or 5200MHz or 4800MHz) Laptop Memory 262-Pin SODIMM, Compatible with Intel Core and AMD Ryzen 7000, Black – CT2K16G56C46S5
Boosts System Performance: 32GB DDR5 RAM laptop memory kit (2x16GB) that operates at 5600MHz, 5200MHz, or 4800MHz to…
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What’s Next
Next steps include ongoing development of SurrealDB to further optimize query planning and indexing, as well as broader testing across diverse workloads and hardware setups. Future benchmarks may explore scalability and performance under different durability settings or in distributed environments. The SurrealDB team plans to release detailed performance guides and continue engaging with the community to validate and expand upon these results.
Ubuntu Linux 24.04 LTS Bootable Live USB Flash Drive for PC/Laptop 64-bit
Ubuntu Linux 24.04 LTS Features: Advanced Threat Protection: Enhanced security features to detect and prevent advanced threats, including…
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
How does SurrealDB’s performance compare to Postgres in real-world applications?
While benchmarks show SurrealDB is approximately 1.5 times faster in write throughput, real-world performance depends on workload specifics, configuration, and deployment environment. SurrealDB’s recent improvements suggest strong potential, but practical results may vary.
What does full durability with fsync enabled mean for performance?
Enabling full durability with fsync ensures every transaction is committed to disk before acknowledgment, which typically results in slower performance compared to cache-based configurations but provides greater data safety and resilience in case of failures.
Are these benchmark results applicable to distributed or cloud deployments?
The benchmarks were conducted on a single, high-performance server. Performance in distributed or cloud environments may differ due to network latency, replication, and other factors. Further testing is needed to assess scalability in such setups.
What are the key internal changes in SurrealDB 3.x that improved performance?
Major internal rewrites include a new query parser, optimized storage layers, and an improved query planner, which collectively reduce per-row decoding overhead and accelerate query execution.
Source: Hacker News
