← zurück zur Übersicht

2026-06-16T20-36-58Z_pgvector-S-B-hybrid-r5

pgvectorStufe S · 10.00 GBhybridStatus: ok

Metriken

Throughput
3.5 QPS
Latenz ⌀
283.72 ms
Latenz p50
240.47 ms
Latenz p95
445.29 ms
Latenz p99
902.39 ms
Recall@1
11.80%
Recall@10
27.72%
Recall@100
32.22%
Precision@10
27.72%
NDCG@10
38.31%

Rohdaten

Konfiguration

Config-Name
pgvector-S-B-hybrid
DB-Version
Image
Index-Typ
ivfflat
Index-Params
{"lists":1000,"probes":10}
Dataset
S · 2.650.000 Vektoren · dim 1024 · Variante B · 10.00 GB
Queries
2.000 (Concurrency 1)

Lauf

Gestartet
16.06.2026, 22:36:58
Beendet
16.06.2026, 22:46:30
Dauer
7772s
Index-Bauzeit
256.48s
Index-Größe
20710.9 MB
CPU ⌀ / Peak
0.53 / 1.21 cores
RAM ⌀ / Peak
7453 / 7552 MB
Disk read (Paging)
168230 MB
Disk write
139 MB
Pod-RAM-Limit
8 GiB
Warmup-Queries
1.000
K8s
v1.31.4+k3s1
Nodes
4

Notizen

measured: in-cluster
decoupled: true
ingested_at: 2026-06-16T18:58:00Z
ingest_config: pgvector-S-B-ivfflat
insert_time_s: 915.51
n_vectors_actual: 2650000
has_metadata: true
mem_limit_gb: 8
pre_run_reset: skipped (decoupled, warmup-discard)
n_queries_executed: 1000
n_warmup: 1000
gt_file: ground_truth_hybrid_alpha_50_ids.npy
mode: measure-only
repeat_group: 2026-06-16T19-59-53Z_pgvector-S-B-hybrid
repeat_index: 5
repeat_total: 6