Launched an enterprise-grade vector database service, rapidly within months
Robust security mechanisms protect data from network threats
Simplifies vector database operations and maintenance, with GKE’s flexible scaling and handling peak demand
Enables cross-region service deployment to meet global enterprises’ high demand for service availability
Increased user retention by offering vector database service on Google Cloud Marketplace, streamlining purchasing
By using Google Cloud, Zilliz achieved rapid development, reduced operational overhead, and secured user data with comprehensive security features.
By using Google Cloud, Zilliz achieved rapid development, reduced operational overhead, and secured user data with comprehensive security features.
Building Zilliz Cloud, a fully managed vector lakebase service, rapidly on Google Cloud
Zilliz is a global leader in vector database and vector lakebase systems, specializing in unstructured data processing for production-ready AI applications. It is the creator of Milvus, a widely adopted open-source vector database with more than 44,000 GitHub stars. In 2023, Zilliz launched Zilliz Cloud, a fully managed, enterprise-grade vector database service built on Milvus to meet the growing demand driven by the generative AI boom. In 2026, Zilliz launched the public preview of Zilliz Vector Lakebase, a major product upgrade to Zilliz Cloud. By early 2026, Zilliz had been adopted by more than 10,000 global enterprises and AI-native startups.
Zilliz chose to build Zilliz Cloud on Google Cloud because of Google Cloud’s leadership in AI, cloud-native infrastructure, and open source technologies. With its strong AI ecosystem and enterprise-grade cloud infrastructure, Google Cloud was a natural choice for powering Zilliz Cloud.
Since the initial stages, Zilliz Cloud has fully adopted a cloud native architecture, leveraging open source technologies, such as Kubernetes and Nginx. The open source technologies and components used in Zilliz Cloud can easily find corresponding alternatives on Google Cloud, such as Google Kubernetes Engine (GKE), Cloud Load Balancing, and Serverless. Zilliz rapidly developed Zillz Cloud and tested 50-60 microservices on Google Cloud in just a few months, enabling the quick launch of its enterprise-grade platform. This helps users quickly launch applications and services based on the latest generative AI technologies.
Google is known as a technology leader in the AI field. Google Cloud has gathered the largest number of AI developers globally. Since Zilliz mainly serves these AI developers and their AI applications, it is natural for us to choose Google Cloud as the infrastructure platform for Zilliz Cloud.
James Luan
CTO, Zilliz
Today, Zilliz Cloud is widely used as the memory layer for agentic AI applications, the retrieval layer for RAG systems, and the AI search infrastructure behind recommendation systems and web search engines. Powered by Zilliz’s internal Cardinal engine, the service can deliver more than 96% recall while keeping indexing latency under 10 milliseconds for billion-scale vector search, helping customers build high-performance AI applications at production scale.
Providing comprehensive security protection while simplifying operations and maintenance
Zilliz provides data services to users in the cloud, so the security of user data needs to be considered from the beginning. Through the Bring Your Own Cloud (BYOC) deployment model that ensures its customers’ data in Zilliz Cloud never leave their own VPCs, Zilliz has easily achieved its security goal with the help of the comprehensive security protection mechanisms on Google Cloud, meeting industry standards such as SOC 2 Type II, ISO 27001, GDPR, and HIPAA.
Google Cloud provides a high-performance operating platform for Zilliz Cloud. Its stability and reliability have helped us provide high-quality vector database and vector lakebase services to users. It is a platform that we can rely on and trust.
James Luan
CTO, Zilliz
For example, Google Cloud Armor is adopted to achieve network security protection. This solution has successfully passed various security tests, reducing security risks. User data is encrypted by managing keys through Cloud Key Management. The user authentication mechanism on Google Cloud is used to implement role-based access control (RBAC). On top of that, the HIPAA-eligible infrastructure of Google Cloud enables Zilliz to ensure data compliance for its customers in the healthcare sector.
For Zilliz’s operation and maintenance team, another advantage of using Google Cloud is the service stability. Reliability and support from Google Cloud for cross-region deployment enable Zilliz to offer users a 99.95% SLA, meeting global enterprises’ high demand for service availability.
With the elastic resources of the cloud and the high scalability of GKE, Zilliz is able to meet its customers’ increasing demand brought by the breakthroughs in generative AI technology. Today, thousands of clusters run on Zilliz Cloud, with only three employees managing operations and maintenance. This high productivity is attributed mainly to the stable and reliable Google Cloud platform, the efficient system architecture of Zilliz Cloud, and the comprehensive operation and maintenance management functions provided by Google Cloud. To ensure the security and reliability of user data, the operation and maintenance team also conducts regular disaster recovery drills, which are very simple and quick on Google Cloud.
Reaching broader users through Google Cloud Marketplace while building AI lakebase
Zilliz Cloud is available on Google Cloud Marketplace, enabling customers to subscribe through a simplified procurement process and unified Google Cloud billing. This helps reduce purchasing complexity while giving Zilliz Cloud an additional channel for customer discovery and sales. Zilliz has observed that customers who purchase through Google Cloud Marketplace are often more familiar with the Google Cloud ecosystem, which can support stronger adoption and retention.
Zilliz is focused on delivering high-performance, cost-efficient vector database and vector lakebase services for production AI applications. As unstructured data volumes grow in the era of AI, cost efficiency has become an important factor for organizations evaluating AI data infrastructure. By building on Google Cloud, Zilliz Cloud can leverage scalable infrastructure to support efficient storage, ingestion, and retrieval at scale. For workloads such as large-scale data import and background processing, Zilliz uses Google Cloud Spot VMs to optimize infrastructure costs while maintaining enterprise-grade performance.
To help customers succeed in the AI era, Zilliz is evolving Zilliz Cloud with Vector Lakebase capabilities. Vector Lakebase combines the low-latency, high-QPS performance of vector databases with the openness, scalability, and economics of data lake architecture. It is designed to provide a unified data foundation for unstructured and multimodal AI data, connecting raw data, embeddings, metadata, indexing, retrieval, and search into a complete AI data lifecycle. This enables developers and enterprises to build more reliable, high-performance AI applications across use cases such as agent memory, RAG, semantic search, recommendation, and web-scale AI search.
At the same time, Zilliz will continue collaborating with Google Cloud to provide powerful underlying technologies for agentic AI and ensure data security for organizations in highly regulated industries, so that the potential use of AI can be securely explored in all sectors.
Our goal is to provide high-performance unstructured data services to global users across all industries. The focus of Google Cloud on agentic AI and AI infrastructure security is in line with our current business goals and will help us bring our leading vector database and vector lakebase services to more users.
James Luan
CTO, Zilliz
Zilliz is a leading provider of open-source and commercial vector database and vector lakebase technologies, with a vision to empower organizations to transform unstructured data into AI-driven insights and innovation. As the creators behind Milvus, the world’s most widely adopted open-source vector database with 44,000+ GitHub stars, Zilliz has been adopted by over 10,000 organizations worldwide for building production AI applications.
Industry: Technology
Location: United States
Products: Google Kubernetes Engine (GKE), Cloud Load Balancing, Serverless, Google Cloud Armor, Cloud Key Management, Google Cloud Marketplace
About Google Cloud partner – Polymeric Cloud
Polymeric Cloud is a Google Cloud partner offering one-stop cloud service and generative AI solutions to enterprises embarking on international expansion. Guided by the mission of “driving innovation and transformation with cloud + generative AI”, the company is committed to empowering enterprises to create a digital future by providing full lifecycle services from cloud technology consulting, architecture design, migration, and O&M management to Google Cloud generative AI application implementation.
