This is the bonus, spoon-fed tutorial in the "Dify x OceanBase seekdb" series. It demonstrates how to deploy Dify—using seekdb as both its metadata database and vector database—on a K8s cluster with just three commands via Helm, and explains how to check Pod status, handle abnormal restarts, and access the service.
Zuoyebang built the data foundation for its AI business on OceanBase. This article details vector database selection, cross-cloud primary-standby tenant replication, read-write splitting, cloud-native proxies, and other multi-cloud architecture designs, and shares large-scale operations experience and cost savings at a scale of 40+ clusters and 20,000+ cores.
An OceanBase engineer offers a developer's take on three open-source AI products—seekdb, PowerRAG, and PowerMem—dissecting the three big challenges of the AI era (multimodal data, context rot, and memory management) and explaining how, under the Data×AI strategy, the three work together to build a complete AI data infrastructure.
In two weekends, the author refactored an enterprise knowledge base from a four-database stack (PostgreSQL + Elasticsearch + Pinecone + Redis) into a single-database solution built on seekdb, OceanBase's open-source AI-native database. Query latency dropped from 120ms to 58ms, saving $450 a month, with complete runnable code and tuning experience included.
Xia Jin, a vector-index engineer on OceanBase, takes a deep dive into how OceanBase / seekdb vector indexes are built and the structure of their five auxiliary tables, analyzes the root causes of slow index builds and high memory usage, and offers optimization methods including adding parallelism, raising the sampling rate, switching to partitioned tables, and rebuilding.
Since 2024, Didi has used OceanBase to replace RocksDB and TokuDB. Using its ride-hailing growth service and core archive database as examples, this article shares large-scale operations experience and new feature practices around hundred-TB data migration, high-concurrency throttling, custom monitoring and alerting, and Binlog Server high availability.
A step-by-step tutorial detailing how to configure OceanBase seekdb as both the metadata database and the vector database in Dify v1.10.1, replacing a multi-component architecture in an integrated way, and quickly building AI applications such as knowledge bases and ChatBots via Alibaba Cloud Bailian models.
Using an intelligent book search app as an example, this article walks you hands-on through data import, vector embedding, and HNSW index creation with seekdb, and implements semantic search, metadata filtering, and RRF hybrid search—giving you a taste of the integrated capabilities of an AI-Native database.
This is a step-by-step tutorial that walks you hands-on through building a RAG (Retrieval-Augmented Generation) system with OceanBase seekdb, covering the full flow of environment setup, configuring three Embedding function types, data import, vector retrieval, and launching a Streamlit app.
After open sourcing its AI-native database seekdb, OceanBase launched a writing contest with prizes, inviting developers to deploy and experience seekdb, or to build AI applications, agents, and system architectures based on seekdb, PowerMem, and PowerRAG, share their hands-on experiences, and win prizes.