References

Unfamiliar with the features of Milvus? Not sure about the meaning of some terms? You are in the right place to find a detailed explanation of the architecture, features, terms, and many other things you may want to know about Milvus.

• Architecture: An anatomy of the architecture, components, data processing workflow of Milvus.

• Benchmark: A benchmark test report covering the performances of data insertion, index building, and vector similarity search in Milvus.

• Boolean Expression Rules: The boolean expression rules and operators for scalar filtering in Milvus.

• Consistency: Milvus provides four levels of consistency for vector similarity search. Learn about the consistency levels and their best-suited scenarios.

• Glossary: The most frequently used terminologies in Milvus.

• In-Memory Replica: The in-memory replica mechanism enables multiple segment replications in the working memory to improve performance and availability.

• Schema: Schema is used to define the properties of a collection and the fields within.

• Similarity Metrics: Milvus supports multiple similarity metrics. Learn how to choose the right metrics that best suit your application.

• System Configurations: A full list of system parameters that can be flexibly configured to best suit your need.

• Time Travel: A detailed explanation of the Time Travel feature and the mechanism behind.

• Vector Index: A full list of the various types of indexes supported by Milvus. Learn about the best-suited scenarios of each index and the configurable index parameters.