Lecture 10 — Capstone: an evidence-based deployment decision¶
Prerequisites: Lectures 01–09. Time: 4–6 hours of integration and review, plus bounded experiments. Outcome: recommend, reject, or defer a specific DeepSeek V4.1 Flash deployment with reproducible evidence.
Project brief¶
Your team needs a service for input-heavy engineering conversations, with a smaller interactive workload sharing capacity. Choose a checkpoint and node configuration. Define input/output distributions, image usage, quality budgets, TTFT/TPOT targets, and a maximum experiment cost before running.
There is no required speedup. A justified rejection of an unsupported or quality-regressing conversion is better engineering than a fast but incorrect demonstration.
Deliverables¶
| Artifact | Minimum content |
|---|---|
| Identity manifest | Model and runtime revisions, conversion provenance, protocol and image preprocessing |
| Compatibility report | Architecture, serialization, loader, kernel, and hardware evidence |
| Capacity model | Per-rank weights/state/workspace, host RAM and disk, startup peak |
| Baseline | Launch, smoke tests, raw request records, quality result |
| Optimization record | At least three hypotheses, including a failed or neutral one if observed |
| Load report | Three repeated runs, latency distributions, failures, goodput, and memory |
| Quality report | Held-out paired results and examined regressions |
| Decision | Recommended operating envelope, unsupported features, limitations, reproduction steps |
Do not fabricate neutral experiments to fill a quota; document what actually happened.
Cost worked example¶
For an illustrative node costing C dollars/hour and useful throughput G output tokens/second:
If G is zero, cost per useful token is undefined/infinite, not zero. Record whether C includes idle time, storage, networking, and operational costs. Keep input-heavy and output-heavy workload accounting separate. Use an actual quoted price only with its date and source.
Review sequence¶
- Reproduce the metadata audit from a clean directory.
- Trace the chosen loader through one expert and one unchanged component.
- Explain the largest capacity-model discrepancy.
- Re-run a quality negative control.
- Recompute metrics from raw records using the supplied summarizer or an equally specified method.
- Explain why the best configuration changes under a different traffic mix.
- Present a one-page decision with uncertainty and an explicit rollout/rollback criterion.
A reviewer should be able to reject your recommendation without disputing what you measured. That is the point of preserving raw evidence.
Analysis-only completion¶
Without the required node or a validated loader, submit the manifest, memory model, format exercises, exact support gap, and a ready-to-run experiment design. Clearly label GPU and full-model fields not run. This earns the analysis milestone; the deployment milestone remains open.
Self-check and rubric¶
Use the roadmap rubric. In addition, the deployment milestone requires actual measured quality and performance for the same checkpoint/runtime tuple. Passing CPU fixtures cannot substitute for those results.
Explain these three distinctions without notes:
- active computation versus resident storage;
- checkpoint-format support versus architecture support;
- raw tokens per second versus quality-preserving goodput.