Content

Applicable to: all users. New feature/specification: Vector Retrieval Service for Milvus introduces a new managed auto scaling policy. Simply set the minimum and maximum node counts for each supported node type. The system then automatically adds or removes nodes within the specified range based on real-time resource load. No scaling schedules, load thresholds, or cooldown periods are required. This policy is suitable for online RAG, multi-tenant retrieval, and other scenarios where traffic fluctuates significantly and is difficult to predict.