The latest research on vector databases from Zilliz, a leading vector database company and the inventor of Milvus, was featured at the 48th International Conference on Very Large Databases (VLDB 2022). The paper, titled “Manu: A Cloud Native Vector Database Management System,” explains Manu, the project name for Milvus 2.0, as a cloud-native, next-generation vector database that implements long-term evolvability, tunable consistency, good elasticity, and high performance. On September 7, Xiaomeng Yi, co-author and Senior Researcher at Zilliz, presented the paper at the conference in an online session on specialized and domain-specific data management.
The design of Manu transcends the philosophies of traditional database management systems (DBMS). It is inspired by the authors’ interaction with the 1200+ industry users of Milvus in the past three years, as vector collections have easily exceeded billion-scale with the development of learning-based embedding models. Manu relaxes data model and consistency constraints in exchange for the elasticity and evolution necessary for a fully managed and horizontally scalable vector database. Manu is also extensively optimized for performance and usability with hardware-aware implementations and support for complex search semantics. The authors evaluated three typical application scenarios to demonstrate Manu’s efficiency, elasticity, and scalability.
The annual VLDB conference is one of the preeminent and most respected venues for the timely dissemination of research and development results in the field of database management. Numerous influential concepts have been introduced there, such as “Big Data,” “Data Science,” and “Cloud” before these terms became widely used. Leaders from the world’s great universities and the developers and engineers for major enterprises come together with students whose research will be the foundation for the next waves of transformation.
The VLDB review board found that Manu/Milvus 2.0 “is one of the few cloud-native specialized vector databases available in the market” and that the team has introduced “a system for efficient similarity search where the data entities consist of attributes and embedded vectors generated by machine learning models.” They continued: “The system is open-sourced [and] has a disaggregated computing and storage architecture, flexible consistency model, time travel capability, and high performance due to the integration of optimizations,” and has “components separated at the functionalities level rather than system level, which make it possible for elastic scaling of each component separately.”
The board further stated: “The paper does a good job of laying out the constraints and design principles needed for a cloud-native vector DBMS, [and provides] real use case scenarios. The future work section also identifies interesting directions.”
“Zilliz has been investing in frontier research since day one,” said Charles Xie, founder and CEO of Zilliz. “Our dedicated R&D department helps us apply the latest research in Milvus to continuously created value for our users. We are glad that our paper was accepted by VLDB ’22, as the industry has once again recognized our technological leadership. We will continue to seek breakthroughs and apply the latest vector database research results in AI, together with institutions, communities, and ecological partners.”
The paper on Manu demonstrates that Zilliz is committed to serving as a primary contributor and committer to Milvus, the popular open source vector database. As a graduate project of the LF AI & Data Foundation, Milvus is purpose-built for processing vector data at scale efficiently and helps organizations create AI applications with ease in various scenarios, including computer vision, NLP, recommendation engines, targeted ads, customized search, smart chatbots, new drug discovery, and much more. The powerful features of Milvus are also incorporated in Zilliz Cloud, a fully managed vector database service that Zilliz now offers in private preview, with the security, reliability, ease of use, and affordability that enterprises require.
Sign up for the free insideAI News newsletter.
Join us on Twitter: https://twitter.com/InsideBigData1
Join us on LinkedIn: https://www.linkedin.com/company/insidebigdata/
Join us on Facebook: https://www.facebook.com/insideAI NewsNOW
