← zurück zur Übersicht

2026-06-16T13-59-46Z_weaviate-S0-tune-hnsw-max-r4

weaviateStufe S0 · 2.00 GBtopkStatus: ok

Metriken

Throughput
169.2 QPS
Latenz ⌀
5.88 ms
Latenz p50
5.87 ms
Latenz p95
6.94 ms
Latenz p99
7.59 ms
Recall@1
97.60%
Recall@10
97.98%
Recall@100
95.33%
Precision@10
97.98%
NDCG@10
98.69%

Rohdaten

Konfiguration

Config-Name
weaviate-S0-tune-hnsw-max
DB-Version
Image
Index-Typ
hnsw
Index-Params
{"M":48,"ef_construction":512,"ef":256}
Dataset
S0 · 550.000 Vektoren · dim 1024 · Variante A · 2.00 GB
Queries
1.000 (Concurrency 1)

Lauf

Gestartet
16.06.2026, 15:59:46
Beendet
16.06.2026, 16:00:26
Dauer
7240s
Index-Bauzeit
3068.94s
Index-Größe
3561.7 MB
CPU ⌀ / Peak
0.46 / 0.63 cores
RAM ⌀ / Peak
2957 / 3655 MB
Disk read (Paging)
0 MB
Disk write
0 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: 2026-06-16T13:57:24Z
ingest_config: weaviate-S0-tune-hnsw-max
insert_time_s: 204.79
n_vectors_actual: 550000
has_metadata: true
mem_limit_gb: 8
pre_run_reset: rollout-restart
timeout_s: 180
gomemlimit_mib: 7000
query_mem_gb: 8
n_queries_executed: 1000
n_warmup: 1000
gt_file: ground_truth_ids.npy
server_latency_ms: {"source":"weaviate_prometheus","metric":"queries_durations_ms","class_name":"Bench","query_type":"get_graphql","count":1000,"count_all_types":1000,"mean_ms":4.798,"windowed":true,"p50_ms":5,"p95_ms":9.5,"p99_ms":9.9}
mode: measure-only
repeat_group: 2026-06-16T13-57-24Z_weaviate-S0-tune-hnsw-max
repeat_index: 4
repeat_total: 6