Iverson GPU

Sample fleet — this is the surface a connected estate produces. Read-only agent, one Helm command.

Fleet

acme-ai — production fleet

4 clusters · 1,728 GPUs · $2.8M/mo

meridian-prod360training-east768inference-west384research216

This month

Recoverable spend

$151K/mo

idle allocated spend at a conservative 55% realization.

Recoverable capacity

164

GPUs back through defragmentation, power-cap fixes and lemon remediation.

Critical exposures

3

correlated-failure findings where one event stops business-critical work.

Loop collisions

2

windows in the last 24h where automation systems fought each other.

Awaiting approval

$146K

simulated savings queued for a human yes — nothing moves without it.

Optimization inbox

What to do Monday morning, ranked. Every action is priced, risk-scored, and reversible — approve it, or simulate it against the governance floor first. Expand a row for the reasoning.

Finding
Value
Risk
Confidence
Action
Reclaim idle H100 reservations
training-east
$86K/mo
Low
96%
Review

61 reserved H100s under 15% utilization for 3+ weeks; right-size the reservation and time-box the holders. Rollback: re-expand the node group.

Redistribute CDU-A workloads
meridian-prod
78% → 57% exposure
Critical
94%
Review

78% of business-critical GPUs share one cooling loop. Move 40 GPUs of standing training onto healthy capacity in other loops (+2.8 kW idle, 5 node moves).

Defragment training pool
training-east
+16 usable GPUs
Low
91%
Review

Free GPUs scattered in sub-gang fragments across 9 nodes; two migration batches recover two 8-GPU gang slots without hardware.

Consolidate inference pool
inference-west
$31K/mo
Medium
88%
Simulate

Off-peak consolidation of the serving fleet saves energy + reservations, but raises single-domain exposure 9 points — simulate against the blast-radius floor before approving.

Cordon 3 lemon nodes
meridian-prod
−54% job-death rate
Low
97%
Review

n23, n35, n29 carry repeat ECC/XID history and keep killing jobs. Cordon, burn-in, re-admit on a clean pass.

Cap rebalancer during the 02:00 window
meridian-prod
$9.4K/mo
Low
93%
Review

The cron rebalancer and cluster-autoscaler fight nightly (see conflict below). One shared migration budget ends the churn.

Control-loop intelligence

Which automation systems are fighting — and what the fight costs.

02:00–03:00meridian-prodConflict detected
$312/window
est. churn cost · ~$9,360/mo if nightly
14
unnecessary migrations
3.7
GPU-hours lost
Event timeline
  1. 02:03Karpenterscaled node pool +2 (underutilization threshold)
  2. 02:07custom rebalancermigrated 11 workloads onto the new nodes
  3. 02:11HPAincreased replicas on chat-serving (queue depth)
  4. 02:13kube-scheduleropened another node for the new replicas
  5. 02:18custom rebalancerconsolidated the same node back down
  6. 02:24Karpenterterminated a node the rebalancer had just filled
  7. 02:31HPAscaled chat-serving back down (queue drained)
Conflicting policies

Karpenter consolidation (underutilized-node TTL 5m) × rebalancer cron (nightly repack)

HPA scale-up (queue depth) × rebalancer consolidation (same pods)

Governed fix

Put both loops under one migration budget with a shared cooldown; the governance weight decides which optimization wins when they disagree.

13:00–14:00inference-westConflict detected
$178/window
est. churn cost · ~$5,340/mo if nightly
9
unnecessary migrations
1.9
GPU-hours lost
Event timeline
  1. 13:04spot replacerrotated 4 spot nodes (price event)
  2. 13:09HPAscaled serving +6 replicas during rotation
  3. 13:16Cluster Autoscaleradded on-demand node for the surge
  4. 13:22spot replacerreclaimed capacity; replicas rescheduled again
Conflicting policies

spot replacement (price threshold) × HPA scale-up (latency SLO)

Governed fix

Gate spot rotation on serving-fleet headroom; price the rotation's churn against the spot discount before acting.

See this for your fleet

Read-only agent, one Helm command. No write access, ever — every action above ships as a reviewed change, not a cluster mutation.

Scheduler stack

or see a full exposure report