Extending the vector database article

[devayani-kv]

[Outline] I would like to add to the existing section of vector databases. The general points I would like to add are:

• How vector databases are different from traditional databases
• Embeddings (I can just link to the already existing section)
• A section on the architecture, namely storage, indexing and query processing
• A primer on Similarity metrics such as Euclidean distance and cosine similarity and an overview of indexing techniques such as HNSW
• A small tutorial on getting started with an opensource database like ChromaDB, covering
  • various chunking techniques
  • importance of chunk sizes and other hyperparameters related to pushing data into the database.