Milvus 2.0: A Vector Database With Cloud-Native Architecture
Talkat KubeCon + CloudNativeCon + Open Source Summit China 2021 · on Thursday, 2021-12-09· at 13:15· Open Source Summit 演讲厅, 虚拟会议 (Virtual)·
summary
Overview of Milvus 2.0, an open-source vector database that targets AI embeddings and unstructured-data analytics by adopting a cloud-native architecture. Covers the new design that decouples read/write and compute/storage, describes Milvus 2.0’s system architecture and main components, and outlines the design considerations and challenges encountered during development.
AI-written summary · about
chapters (22)
- 0:00Intro
- 1:36Unstructured Data Trend
- 2:57Unstructured Data Growth
- 3:25Al-driven Data Search and Analytics
- 5:17The Big Picture
- 6:48Milvus is an LF AI & Data Graduated Project
- 7:13Milvus Community
- 8:09Milvus is used in almost every industry
- 9:36Milvus is used for almost every type of unstructured data
- 11:08System Overview
- 14:08Features
- 16:58Log Sequence Pub-sub as System Backbone
- 20:40Access Layer
- 21:51Coordinator Layer
- 23:09Worker Layer
- 24:38Storage Layer
- 25:13Combine Streaming and Batching
- 26:57The Bazaar Architecture
- 28:41History of Milvus
- 29:34Milvus Insight
- 30:00Milvus CLI
- 30:12K8s Operator
related talks
29mThe Evolution of Milvus: A Cloud-Native Vector Database
37mMilvus: Scaling Vector Data Solutions with Gen AI- 37mMilvus: Scaling Vector Data Solutions with Gen AI
40mEvolution of Milvus Cloud-Scalable Vector Database
30mMilvus Vector Database: Unlocking the Future of Open Source Vector Databases
36mMilvus: Scaling Vector Data Solutions with Gen AI
35mMilvus 2.0: Building the World's Most Advanced Open-source Vector Database
51mUnleashing the Potential of Cloud-Native Vector Databases