ABN Techweek: Vector DB

In ABN Techweek this week, we look at Vector DB and why they are useful. Lets dive in.

Vector databases are so hot right now, but what is a Vector DB?

The diagram below shows a comparison between a vector database and other types of databases.

🔹 A vector database indexes and stores vector embeddings for fast retrieval and similarity search, with capabilities like CRUD operations, metadata filtering, and horizontal scaling.

🔹 Recent advances in AGI (Artificial General Intelligence) have made vector databases so popular.

🔹 A vector database stores high-dimensional vectors extracted from various unstructured data, like audio, video, image, and text. Then we can calculate the similarity among unstructured data. Typical use cases include:

– finding similar images or text
– recommending similar products
– detecting abnormalities
– temporarily store embeddings for large amounts of input

🔹 There has been a great deal of funding raised by vector database companies:
– Pinecone: $138 million
– Milvus: $113 million
– Weaviate: $67.7 million
– Chroma: $20 million
– Qdrant: $9.8 million

Over to you: Redis, ElasticSearch, and PostgreSQL support vector data processing. Are specialized vector databases necessary?

AbnAsia.org Software. Faster. Better. More Reliable. +84945924877 (Asia# Mobile, WhatsApp, Telegram, Viber, Zalo); +16699996606 (US# Mobile, WhatsApp, Telegram) [email protected]