← zurück zur Übersicht

2026-06-18T07-06-51Z_pgvector-S1-ivfflat-throughput-c8-r4

pgvectorStufe S1 · 5.00 GBbatchStatus: ok

Metriken

Throughput
21.0 QPS
Latenz ⌀
379.65 ms
Latenz p50
376.15 ms
Latenz p95
471.08 ms
Latenz p99
538.89 ms
Recall@1
67.70%
Recall@10
63.70%
Recall@100
55.71%
Precision@10
63.70%
NDCG@10
73.65%

Rohdaten

Konfiguration

Config-Name
pgvector-S1-ivfflat-throughput-c8
DB-Version
Image
Index-Typ
ivfflat
Index-Params
{"lists":1200,"probes":10}
Dataset
S1 · 1.350.000 Vektoren · dim 1024 · Variante A · 5.00 GB
Queries
2.000 (Concurrency 8)

Lauf

Gestartet
18.06.2026, 09:06:51
Beendet
18.06.2026, 09:08:31
Dauer
7300s
Index-Bauzeit
162.47s
Index-Größe
10556.3 MB
CPU ⌀ / Peak
0.36 / 0.43 cores
RAM ⌀ / Peak
7447 / 7460 MB
Disk read (Paging)
16477 MB
Disk write
80 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: 2026-06-18T07:01:29Z
ingest_config: pgvector-S1-ivfflat
insert_time_s: 546.25
n_vectors_actual: 1350000
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_ids.npy
mode: measure-only
repeat_group: 2026-06-18T07-01-30Z_pgvector-S1-ivfflat-throughput-c8
repeat_index: 4
repeat_total: 6