Vector Database
Vector Database — БД, оптимизированная для хранения + поиска векторов (embeddings) через ANN (Approximate Nearest Neighbor) алгоритмы. HNSW, IVF, DiskANN — common indexing. Поддерживает hybrid search (vector + keyword), metadata filtering, upsert/delete. 2026 leaders: Qdrant (Rust open-source, fastest), Pinecone (managed, expensive), Weaviate (open hybrid search), Milvus (CNCF), pgvector (Postgres extension, простой).
Ниже: подробности, пример, смежные термины, FAQ.
Подробности
- Scale: миллиарды vectors на single node с SSD disk-based indexes
- Index types: HNSW (fastest recall) vs IVF (cheaper memory) vs DiskANN (huge scale)
- Hybrid search: dense vector + sparse BM25 + metadata filter
- Cloud managed: Pinecone, Qdrant Cloud, Weaviate Cloud, Zilliz (Milvus)
- Self-host: Qdrant / Weaviate / Milvus Docker image, pgvector extension в Postgres 14+
Пример
# Qdrant: create collection + insert + search
curl -X PUT http://localhost:6333/collections/docs \
-H 'Content-Type: application/json' \
-d '{"vectors": {"size": 1536, "distance": "Cosine"}}'
# Insert
curl -X PUT http://localhost:6333/collections/docs/points \
-d '{"points": [{"id": 1, "vector": [0.1, 0.2, ...], "payload": {"text": "..."}}]}'
# Search top-5
curl -X POST http://localhost:6333/collections/docs/points/search \
-d '{"vector": [0.1, 0.2, ...], "limit": 5}'Смежные термины
Больше по теме
Часто задаваемые вопросы
pgvector vs dedicated vector DB?
pgvector: simplicity (ваш уже Postgres), до ~1M vectors. Dedicated: лучше при >10M, hybrid search, HA. Начинайте с pgvector, migrate если upgrade нужен.
Qdrant vs Pinecone?
Qdrant: open source, self-host, Rust быстрый, $0 cost. Pinecone: managed, no ops, $70+/mo. Enterprise features: Qdrant Cloud или Pinecone.
Как monitor?
<a href="/ping">Enterno Ping</a> для порта 6333 (Qdrant). <a href="/monitors">Monitors</a> для /health endpoint.