As a cloud-native vector database, Milvus separates storage and computation by design. To enhance elasticity and flexibility, all components in Milvus are stateless.
Milvus Architecture Overview: Milvus adopts a shared-storage architecture featuring storage/computing disaggregation and scalability for its computing nodes.
Storage/Computing Disaggregation: Milvus comprises four layers that are mutually independent in terms of scalability and disaster recovery.
Main Components: Milvus standalone includes three components while Milvus cluster includes eight microservice components and three third-party dependencies.
Data Processsing: A detailed description of the implementation of data insertion, index building, and data query in Milvus.