👻 Ghostfleet run report

🖥️ 1000 nodes · 🎮 8000 GPUs · 📦 10500 pods · snapshot 2026-07-19T08:34:01Z

📋 Cluster shape

🖥️ Nodes
1000
🎮 GPUs
8000
📦 Pods
10500
🗄️ etcd DB size
—MB
⚡ Scheduler throughput
13.5pods/s (5m avg)
⏱️ Scheduling e2e p99
69.675s

🌐 API request p99 latency by verb

PATCH
763.03 ms
DELETE
678.50 ms
GET
105.27 ms
LIST
2860.00 ms
POST
60000.00 ms
PUT
304.43 ms
APPLY
no data
Table view
Verbp99
PATCH763.03 ms
DELETE678.50 ms
GET105.27 ms
LIST2860.00 ms
POST60000.00 ms
PUT304.43 ms
APPLYno data

SLO: mutating < 1s · read-only namespaced < 5s · read-only cluster-scoped < 30s

🗄️ etcd request p99 latency by operation

list
60000.00 ms
listWithCount
1980.00 ms
update
60000.00 ms
delete
98.17 ms
create
1366.68 ms
get
171.95 ms
Table view
Operationp99
list60000.00 ms
listWithCount1980.00 ms
update60000.00 ms
delete98.17 ms
create1366.68 ms
get171.95 ms

⚖️ Priority & Fairness

✓ No APF rejections or queue wait observed in this window.

👀 Watchers & inflight requests

No watcher data in this window.

Inflight requests
KindCount

Generated from results/20260719-103101/ — raw PromQL snapshots in that folder are the source of truth.