Zilliz Cloud Serverless, built on the open-source Milvus, offers an auto-scaling vector database designed to facilitate effortless application building, especially for GenAI applications. It features automated tiered storage, which optimizes data placement across DRAM, SSD, and object storage, ensuring high performance for frequently accessed data while reducing costs for less frequently used information. This approach makes it significantly more cost-efficient—up to 50 times—compared to traditional in-memory vector databases.
Zilliz Cloud Serverless eliminates the need for database setup, maintenance, and scaling concerns, allowing developers to focus on building applications. It dynamically adjusts compute resources to match application demands, ensuring that users only pay for what they use, making it ideal for fluctuating workloads. Additionally, it offers data portability, allowing users to transfer data to dedicated clusters or open-source Milvus as their needs evolve.
The platform supports high-performance vector search, low latency with high recall, and hybrid search capabilities, enabling simultaneous querying across multiple vector fields. It also provides various similarity metrics and tunable consistency levels to enhance classification, clustering performance, and data accuracy. The component-based architecture allows for effortless horizontal scaling, ensuring peak performance and efficiency regardless of workload fluctuations.
Pricing
Zilliz offers flexible pricing options for its fully-managed vector database service, Zilliz Cloud, catering to various team budgets. A cost calculator is available to estimate expenses, and a free tier is provided for initial use.