Meet OceanBase AI Database, the unified database for operational data, real-time analytics, and AI. Explore ->
Meet OceanBase AI Database, the unified database for operational data, real-time analytics, and AI. Explore ->
Rethinking the Database for the AI Era
AI is changing what enterprises need from databases. Learn how OceanBase AI Database combines lakebase architecture, multimodal data, hybrid search, and agent-friendly capabilities for the AI era.
Beyond High Availability: How OceanBase Keeps Data Correct
How OceanBase goes beyond high availability with physical and logical verification, LSM-based compaction, and recovery to keep data correct.
Committing a 1 TB Transaction as Fast as 1 MB — How MaLT Makes Commit, Rollback, and Recovery Constant-Time
MaLT embeds transaction metadata into LSM-tree, making commit, rollback, and recovery constant-time regardless of size. Bulk import of 2.5M rows speeds up 23.9%; recovery stays under 25s.
Inside OceanBase Migration, Replication, and Rebuild: Moving Data Without Taking the Database Offline
OceanBase uses background DR tasks (migrate, add, type-transform, rebuild) to keep tenant locality compliant, all online without pausing writes, observable via task views.
Zero Data Loss, Fast Failover: How OceanBase Uses Multi-Paxos for High Availability
How OceanBase uses Multi-Paxos, log streams, replica types, and leader election to achieve zero data loss and fast failover.
OceanBase Failover Routing: How OBProxy and Leader Scheduling Carry RTO < 8s Toward the Client
RTO < 8s on the cluster isn't the same as RTO < 8s to the application. Trace how OBProxy's Location Cache, route feedback, and bounded retry close the gap.
Why a new AI-native database trended on GitHub
seekdb, a MySQL-compatible state store for AI agents, delivers hybrid search in one SQL, 1,523 QPS streaming with 1.1× P99 jitter, and kernel-level Fork/Diff/Merge sandboxes for safe agent exploration.
Why Streaming Workloads Break Most Vector Databases: A Benchmark of 5 Popular Engines
Streaming benchmark shows traditional bulk-load tests miss agent needs. seekdb's fixed dual-index keeps P99 jitter at 1.1×.
From Neurons to Code: The Forgetting Design Behind PowerMem
Nature designed memory and forgetting together. We want to translate that design into code that can be configured and tuned.
Locality in OceanBase: How Replica Placement Becomes an Availability Contract
Learns how locality in OceanBase turns a DR topology into something the cluster enforces - replica counts, replica types (F/R/C), and zone placement.
OceanBase Blog
Product insights, engineering deep dives, and real-world use cases from the OceanBase team.
AI is changing what enterprises need from databases. Learn how OceanBase AI Database combines lakebase architecture, multimodal data, hybrid search, and agent-friendly capabilities for the AI era.
Rethinking the Database for the AI Era
AI is changing what enterprises need from databases. Learn how OceanBase AI Database combines lakebase architecture, multimodal data, hybrid search, and agent-friendly capabilities for the AI era.
Beyond High Availability: How OceanBase Keeps Data Correct
How OceanBase goes beyond high availability with physical and logical verification, LSM-based compaction, and recovery to keep data correct.
Committing a 1 TB Transaction as Fast as 1 MB — How MaLT Makes Commit, Rollback, and Recovery Constant-Time
MaLT embeds transaction metadata into LSM-tree, making commit, rollback, and recovery constant-time regardless of size. Bulk import of 2.5M rows speeds up 23.9%; recovery stays under 25s.
Inside OceanBase Migration, Replication, and Rebuild: Moving Data Without Taking the Database Offline
OceanBase uses background DR tasks (migrate, add, type-transform, rebuild) to keep tenant locality compliant, all online without pausing writes, observable via task views.
Zero Data Loss, Fast Failover: How OceanBase Uses Multi-Paxos for High Availability
How OceanBase uses Multi-Paxos, log streams, replica types, and leader election to achieve zero data loss and fast failover.
OceanBase Failover Routing: How OBProxy and Leader Scheduling Carry RTO < 8s Toward the Client
RTO < 8s on the cluster isn't the same as RTO < 8s to the application. Trace how OBProxy's Location Cache, route feedback, and bounded retry close the gap.
Why a new AI-native database trended on GitHub
seekdb, a MySQL-compatible state store for AI agents, delivers hybrid search in one SQL, 1,523 QPS streaming with 1.1× P99 jitter, and kernel-level Fork/Diff/Merge sandboxes for safe agent exploration.
Why Streaming Workloads Break Most Vector Databases: A Benchmark of 5 Popular Engines
Streaming benchmark shows traditional bulk-load tests miss agent needs. seekdb's fixed dual-index keeps P99 jitter at 1.1×.
From Neurons to Code: The Forgetting Design Behind PowerMem
Nature designed memory and forgetting together. We want to translate that design into code that can be configured and tuned.
Locality in OceanBase: How Replica Placement Becomes an Availability Contract
Learns how locality in OceanBase turns a DR topology into something the cluster enforces - replica counts, replica types (F/R/C), and zone placement.
Designing Disaster Recovery with OceanBase: A Failure-Domain Playbook
How to choose the right OceanBase disaster recovery architecture by failure domain - Server, Zone/IDC, Region, or full cluster - with a decision matrix.
Why Your Vector Database Benchmark Is Wrong for AI Agents
Under streaming AI workloads, vector databases see high P99 jitter (1.1×–10.3×) under concurrency. seekdb v1.3.0’s fixed delta+snapshot HNSW avoids this, delivering 22× QPS and 19× P99 gains over prior version.