QDrant. Bases de dades semàntiques

QDrant

• [[1]]

docker pull qdrant/qdrant

docker run -p 6333:6333 qdrant/qdrant
...
Access web UI at http://localhost:6333/dashboard

Qdrant is now available at:

    REST API: http://localhost:6333

    Web UI: http://localhost:6333/dashboard

Segueixo la guia del quickstart:

Afegim una nova col·lecció:

PUT collections/star_charts

{

  "vectors": {

    "size": 4,

    "distance": "Dot"

  }

}

Afegim punts a la col·lecció:

PUT collections/star_charts/points

{
  "points": [
    {
      "id": 1,
      "vector": [0.05, 0.61, 0.76, 0.74],
      "payload": {
        "colony": "Mars"
      }
    },
    {
      "id": 2,
      "vector": [0.19, 0.81, 0.75, 0.11],
      "payload": {
        "colony": "Jupiter"
      }
    },
    {
      "id": 3,
      "vector": [0.36, 0.55, 0.47, 0.94],
      "payload": {
        "colony": "Venus"
      }
    },
    {
      "id": 4,
      "vector": [0.18, 0.01, 0.85, 0.80],
      "payload": {
        "colony": "Moon"
      }
    },
    {
      "id": 5,
      "vector": [0.24, 0.18, 0.22, 0.44],
      "payload": {
        "colony": "Pluto"
      }
    }
  ]
}

Step 3: Run a search query

POST collections/star_charts/points/search
{
  "vector": [0.2, 0.1, 0.9, 0.7],
  "limit": 3,
  "with_payload": true
}

resultat:

{
  "result": [
    {
      "id": 4,
      "version": 1,
      "score": 1.362,
      "payload": {
        "colony": "Moon"
      }
    },
    {
      "id": 1,
      "version": 1,
      "score": 1.273,
      "payload": {
        "colony": "Mars"
      }
    },
    {
      "id": 3,
      "version": 1,
      "score": 1.208,
      "payload": {
        "colony": "Venus"
      }
    }
  ],
  "status": "ok",
  "time": 0.000833372
}

• • Persistència**: que és guardin les dades

docker run -p 6333:6333 -p 6334:6334 \
    -v $(pwd)/qdrant_storage:/qdrant/storage:z \
    -v $(pwd)/qdrant_snapshots:/qdrant/snapshots:z \
    -v $(pwd)/config/production.yaml:/qdrant/config/production.yaml:z \
    qdrant/qdrant

primer he de crear el fitxer production.yaml amb la configuració mínima com es comenta en el tutorial.

2026-06-03T15:56:36.213394Z ERROR qdrant: Error while starting REST server: TLS file could not be opened: ./tls/cert.pem

Error: TLS file could not be opened: ./tls/cert.pem

Caused by:

    failed to open file `./tls/cert.pem`: No such file or directory (os error 2

-> en el fitxer de configuració he de deshabilitar tls

Ja puc entrar però ara em demana una api key:

This instance of Qdrant might be protected by an API Key. If so, please enter your API Key to continue.

la api key es deshabilita en el fitxer de configuració

Ja funciona el tutorial, i ara ja tinc persistència: és a dir, puc tancar el docker i tornar-lo a obrir, i les dades encara hi són. Recordar que tenim dos volums que s'han creat.

Per ex:

GET collections/terraforming

Puc executar la rest api en **línia de comanda**. Per ex:

curl -X GET \
  "http://localhost:6333/collections/terraforming" \
  -H "Content-Type: application/json"

Resultat:

{"result":{"status":"green","optimizer_status":"ok","indexed_vectors_count":0,"points_count":6,"segments_count":8,"config":{"params":{"vectors":{"size":4,"distance":"Dot"},"shard_number":1,"replication_factor":1,"write_consistency_factor":1,"on_disk_payload":true},"hnsw_config":{"m":16,"ef_construct":100,"full_scan_threshold":10000,"max_indexing_threads":0,"on_disk":false},"optimizer_config":{"deleted_threshold":0.2,"vacuum_min_vector_number":1000,"default_segment_number":0,"max_segment_size":null,"memmap_threshold":null,"indexing_threshold":10000,"flush_interval_sec":5,"max_optimization_threads":null,"prevent_unoptimized":null},"wal_config":{"wal_capacity_mb":32,"wal_segments_ahead":0,"wal_retain_closed":1},"quantization_config":null},"payload_schema":{"life":{"data_type":"bool","points":6}},"update_queue":{"length":0}},"status":"ok","time":0.001508578}