← Back to the chapter · Previous detail
Technical detail · Hardware Accelerators for Edge AI
Keep the decision record tied to evidence
13.11 Implementation Sketch
The sketch gives two points each for supported operators, sufficient memory headroom, measured latency within the deadline, and both thermal validation and rollback. Its returned risks make missing evidence visible; a high score does not waive a failed requirement.
def score_candidate(candidate, requirements):
score = 0
risks = []
if candidate.supports(requirements.model_operators):
score += 2
else:
risks.append("unsupported model operators")
if candidate.memory_headroom >= requirements.memory_headroom:
score += 2
else:
risks.append("insufficient memory headroom")
if candidate.measured_latency_ms <= requirements.deadline_ms:
score += 2
else:
risks.append("misses end-to-end latency deadline")
if candidate.thermal_test_passed and candidate.update_path.has_rollback:
score += 2
else:
risks.append("thermal or update validation incomplete")
return {"candidate": candidate.name, "score": score, "risks": risks}
13.11.1 Evidence Ledger
| Check | Evidence To Record | Failure Signal |
|---|---|---|
| Operator placement | Runtime/compiler logs for the converted graph | Unsupported layer or CPU fallback |
| Compute demand | MACs per inference times inferences per second | Missed deadline despite good peak metrics |
| Memory headroom | Weights, activations, input buffers, tensor arena, runtime | Allocation failure, swapping, or smaller batch/window than required |
| Sustained operation | Power, thermals, and enclosure run under expected duty cycle | Throttling, resets, battery miss, or heat limit |
| Lifecycle support | Model/runtime update path, monitoring, rollback | Manual updates or unobservable fleet state |