2026-06-18T18-11-14Z_pgvector-S-tune-hybrid-a30-r4
pgvectorStufe S · 10.00 GBhybridStatus: ok
Metriken
Throughput
3.3 QPS
Latenz ⌀
304.19 ms
Latenz p50
267.93 ms
Latenz p95
452.07 ms
Latenz p99
781.05 ms
Recall@1
5.30%
Recall@10
4.83%
Recall@100
7.49%
Precision@10
4.83%
NDCG@10
5.79%
Rohdaten
summary.json ↗Vollständige Run-Metrikenhttps://raw.githubusercontent.com/unrealshape/bachelor-db-benchmark/main/results/2026-06-18T18-11-14Z_pgvector-S-tune-hybrid-a30-r4/summary.jsonconfig.json ↗Eingesetzte Konfigurationhttps://raw.githubusercontent.com/unrealshape/bachelor-db-benchmark/main/results/2026-06-18T18-11-14Z_pgvector-S-tune-hybrid-a30-r4/config.jsonDatensatz · 10.00 GB ↗Generator + Bauanleitung im RepoRun-Ordner auf GitHub ↗Beide Rohdateien im Verzeichnis
Konfiguration
- Config-Name
- pgvector-S-tune-hybrid-a30
- DB-Version
- —
- Image
- —
- Index-Typ
- ivfflat
- Index-Params
- {"lists":2400,"probes":10}
- Dataset
- S · 2.650.000 Vektoren · dim 1024 · Variante A · 10.00 GB
- Queries
- 1.000 (Concurrency 1)
Lauf
- Gestartet
- 18.06.2026, 20:11:14
- Beendet
- 18.06.2026, 20:21:27
- Dauer
- 7813s
- Index-Bauzeit
- 276.68s
- Index-Größe
- 20710.9 MB
- CPU ⌀ / Peak
- 0.49 / 1.43 cores
- RAM ⌀ / Peak
- 7449 / 7482 MB
- Disk read (Paging)
- 175941 MB
- Disk write
- 20 MB
- Pod-RAM-Limit
- Default
- Warmup-Queries
- 1.000
- K8s
- v1.31.4+k3s1
- Nodes
- 4
Notizen
measured: in-cluster decoupled: true ingested_at: 2026-06-18T16:40:45Z ingest_config: pgvector-S-ivfflat insert_time_s: 1021.04 n_vectors_actual: 2650000 has_metadata: true mem_limit_gb: null pre_run_reset: skipped (decoupled, warmup-discard) n_queries_executed: 1000 n_warmup: 1000 gt_file: ground_truth_hybrid_alpha_30_ids.npy mode: measure-only repeat_group: 2026-06-18T17-40-15Z_pgvector-S-tune-hybrid-a30 repeat_index: 4 repeat_total: 6