experiment: add workload regime taxonomy

.research/experiments/frontier-workload-regime-taxonomy-20260720.md

@@ -0,0 +1,160 @@
+# Frontier workload-regime taxonomy
+
+- Date: 2026-07-20
+- Status: proposed; awaiting review before workload generation or GPU runs
+- Scope: explain when Frontier preserves the real-system config ranking, rather than merely comparing Fixed with Trace
+
+## Claim under test
+
+Frontier reliability is controlled by three quantities:
+
+1. the latency-model residual between simulator and real execution;
+2. the closed-loop gain from timing to scheduler state (batch, MoE routing, CUDA-graph bucket, MNS occupancy, admission/KV pressure);
+3. the real decision margin between configurations.
+
+For a config pair `a,b`, define
+
+```text
+D_real(a,b) = log L_real(a) - log L_real(b)
+delta(a,b)  = [log L_sim(a)-log L_real(a)]
+              - [log L_sim(b)-log L_real(b)]
+slack(a,b)  = sign(D_real) * [D_real + delta]
+```
+
+`slack < 0` means the simulator reverses the real pairwise ordering. The primary hypothesis is that reversals occur when simulator and real execution land on different sides of a scheduler-state knee, or when the real decision margin is too small to absorb the differential residual. `Fixed` and `Trace` are not themselves the causal classes.
+
+## Existing evidence motivating the experiment
+
+- Q30 Trace-PD preserves all six objective winners, but many pairwise residuals oppose the real winner. Its success is therefore often margin protection, not zero residual.
+- Q235 Trace-PD preserves TTFT/TPOT winners but misses E2E p90 by 6.2%; Trace is not universally safe.
+- Q30/Q235 Fixed-PD decode objectives show negative minimum signed slack and 13--37% regret.
+- In Q30 low-load Fixed-PD, Frontier's batch-1 TP ordering is correct, while the closed-loop simulator increases TP4's effective batch and changes the MoE cost enough to reverse the ordering. This identifies a concrete state knee, but does not yet establish a general rule.
+
+## Workload families
+
+All comparisons use the same request multiset where applicable, the same total observation window, and the same normalized offered decode load
+
+```text
+rho = request_rate * E[output_tokens] / measured_reference_decode_capacity.
+```
+
+This avoids equating equal request rates with equal load.
+
+| ID | Shape / request lengths | Arrival process | Prefix/session state | Isolated effect |
+|---|---|---|---|---|
+| W0 | short fixed `2048 -> 128` | uniform | off | known low-residence failure anchor |
+| W1 | trace-mean fixed ISL/OSL | uniform | off | homogeneous baseline |
+| W2 | trace-mean fixed ISL/OSL | trace timestamps | off | arrival burst only |
+| W3 | exact trace ISL/OSL multiset | uniform | off | length heterogeneity only |
+| W4 | exact trace ISL/OSL multiset | trace timestamps | off | length + burst |
+| W5 | exact trace prompts/ISL/OSL | uniform | exact prefix/session identity | prefix state without burst |
+| W6 | exact trace prompts/ISL/OSL | trace timestamps | exact prefix/session identity | full production trace |
+
+Prefix is intentionally a nested factor: enabling a synthetic prefix graph on fixed identical requests would introduce a different workload rather than isolate production prefix reuse. Therefore this is not presented as a full `2^3` factorial.
+
+## Load sweep and expected patterns
+
+Simulator discovery sweep: `rho in {0.05, 0.25, 0.50, 0.90, 1.20}`. The points mean deep low load, light batching, moderate batching, capacity knee, and overload; their request rates are derived independently for every workload family.
+
+| Pattern | Observable state | Prediction for Frontier |
+|---|---|---|
+| P1 singleton-linear | real and sim stay below the first batch/graph knee | works if the batch-1 operator ordering is correct |
+| P2 knee-straddling | real and sim occupy opposite sides of a batch/MoE/graph/MNS knee | fails systematically; Fixed-PD is the current example |
+| P3 same-side batched | both systems cross the same knee and remain below admission pressure | works if batch-conditioned operator ordering is correct |
+| P4 capacity/admission aligned | both systems are governed by the same capacity bottleneck | TTFT/config winner may work despite large absolute error; E2E/MNS can remain fragile |
+| P5 heterogeneity-smoothed | broad lengths reduce coherent threshold occupancy at matched `rho` | may work; this is a hypothesis, not an established explanation |
+| P6 burst-sensitive | same request multiset, but transient queue/MNS occupancy differs | mean ranking may work while TTFT/E2E tail ranking fails |
+| P7 prefix-state-sensitive | hit/eviction and reused-token distributions differ | TTFT ranking fails unless prefix-state transitions are modeled; decode TPOT may remain stable |
+| P8 decision-boundary | real config margin is comparable to run variance/residual | fragile; an exact winner match is not reliable evidence |
+
+## Hypotheses and distinguishing tests
+
+### H1: state-regime hypothesis (primary)
+
+I believe config-ranking failures occur when the latency residual moves a workload across a scheduler-state knee, because the residual is then amplified into a different batch/resource trajectory. I will verify this by checking whether signed-slack zero crossings co-locate with measured real/simulator state-knee crossings.
+
+### H2: heterogeneity-smoothing hypothesis
+
+I believe length heterogeneity can reduce coherent threshold amplification, because requests reach scheduler boundaries at dispersed times. I will verify it with W1 vs W3 and W2 vs W4 at matched `rho`, requiring a smaller real/sim state-distribution gap rather than merely a correct winner.
+
+### H3: bottleneck/margin-protection alternative
+
+Trace success may instead be explained entirely by a large real decision margin or a shared capacity bottleneck. This hypothesis wins over H2 if W3/W4 do not reduce state-distribution error after matching load and margin, while ranking correctness remains predicted by margin alone.
+
+### H4: burst and prefix are independent failure channels
+
+I believe arrival bursts primarily affect waiting/admission and tail TTFT/E2E, whereas prefix mismatch primarily affects prefill/TTFT state. I will verify this with W1/W2, W3/W4, and W3/W5 paired comparisons.
+
+## Configuration and model scope
+
+Discovery uses Qwen30B because its 12-cell `TP x MNS` surface already has simulator and real anchors:
+
+- TP: `{1, 2, 4}`
+- MNS: `{8, 32, 64, 128}`
+- objectives: mean/p90 TTFT, TPOT, E2E
+
+Qwen235B is a held-out confirmation, not pooled into discovery:
+
+- existing four feasible TP/MNS configurations;
+- only the workload/load patterns that discriminate H1--H4 after Q30 converges.
+
+## Measurements
+
+End-to-end:
+
+- completed/failed requests and achieved request/token rate;
+- TTFT, TPOT, E2E mean/p50/p90/p95;
+- config regret, pairwise agreement, signed decision slack;
+- run-to-run winner stability.
+
+Closed-loop state:
+
+- prefill/decode batch-size histograms and time-weighted batch;
+- Running/Waiting distributions and admission delay;
+- MNS active-token occupancy and KV/context pressure;
+- CUDA-graph bucket residency and fallback frequency;
+- prefix hit/reused-token/eviction distributions for W5/W6.
+
+## Decision rules
+
+A workload/load region is:
+
+- **reliable** if regret is at most 5%, pairwise agreement is at least 0.8 at two adjacent load points, and the winner is stable across confirmation trials;
+- **fragile** if regret is at most 5% but the real margin overlaps run uncertainty, or a small rate/timing perturbation changes the winner;
+- **failed** if regret exceeds 5% or a decision-critical pair has negative signed slack;
+- **mechanistically explained by H1** only if the ranking transition co-locates with an observed state-regime transition. Correlation with the Fixed/Trace label is insufficient.
+
+H2 is supported only if the heterogeneous member of a matched pair reduces state-distribution error and shifts the failure boundary in repeated trials. A correct winner alone does not support smoothing.
+
+## Execution plan after review
+
+1. Materialize W0--W6 with one manifest recording request multiset, arrival timestamps, prefix identity, rate contract, and hashes.
+2. Run the simulator sweep across `rho` and the Q30 surface; emit a per-stage state ledger.
+3. Select real-machine pilot points only around the predicted knees plus one safe-side control. Use guard configs `TP1/MNS64`, `TP4/MNS8`, and `TP4/MNS64`; add `TP2/MNS32` only if the transition is not bracketed.
+4. Use only `dash1`, `dash2`, `dash3`, and `dash4`, each verified as an 8×H20 host. `dash0` is excluded from probing, synchronization, and execution. Pin one independent experiment group to each host so at most four groups run in parallel; do not split one trial across hosts.
+5. Run one pilot trial per selected point. Confirm only hypothesis-discriminating points with three fresh-server trials and rotated order.
+6. Apply the resulting classifier unchanged to the Q235 held-out cases.
+
+Provisional four-way allocation after the simulator identifies the discriminating points:
+
+| Host | Experiment group | Primary contrast |
+|---|---|---|
+| dash1 | homogeneous controls | W0/W1 across safe side and first knee |
+| dash2 | arrival effect | W1 vs W2 and W3 vs W4 |
+| dash3 | length heterogeneity | W1 vs W3 and W2 vs W4 |
+| dash4 | prefix/full trace | W3 vs W5 and W4 vs W6 |
+
+The groups are logical queues, not permanent ownership: if a host probe fails, that host is excluded and its group waits or moves to another permitted idle host. Cross-host latency values are not pooled until a common canary config verifies that host effects are within run uncertainty.
+
+No GPU run is authorized by this card yet. The review decision is whether the workload decomposition and decision rules are sufficient to implement the materializer and launch Phase 1.
+
+## Expected figure
+
+The accompanying mock figure is schematic, not data. Panel A shows the state knee that real and simulator trajectories may cross at different loads. Panel B shows the corresponding minimum signed decision slack; a negative value denotes a ranking reversal. The claim is supported only if measured zero crossings and state knees align across workload families.
+
+## Risks and controls
+
+- Equal `rho` does not guarantee equal prefill pressure; report both prefill and decode offered work and stratify if necessary.
+- Full-trace overload can collapse all configs to similarly poor latency. Such points identify a capacity-limited region but cannot validate fine-grained ranking.
+- MNS ties and censored/failed requests can create false winners; exclude invalid cells before calculating regret and report the exclusion.
+- One trace cannot establish generality. The initial result is a mechanism boundary for this trace/model/hardware, followed by held-out Q235 validation.

.research/ongoing.md

@@ -5,14 +5,18 @@
 > **2026-07-19 update：** Qwen235 Fixed-PD 的错误排序在 exact real state composition 下已经翻正，主因是 simulator closed-loop batch state，而不是 collective。Qwen30 Fixed-PD 的 56--58% TTFT regret 也已定位：Frontier 将 decode service time 高估 4--8×，使 TP4 的 modeled concurrency 越过 MNS admission cap并产生虚假排队；去掉该等待后 Frontier 与真机都判定 TP4 topology 更快。详见 [`experiments/qwen30-fixed-pd-ttft-admission-diagnosis-20260719.md`](experiments/qwen30-fixed-pd-ttft-admission-diagnosis-20260719.md)。
 >
 > **2026-07-20 update：** 对全部 14 个 frozen case surface（70 个 case×objective）做了统一的 margin-vs-residual 分解与方向化机制普查（[`experiments/frontier-split-rootcause-s0-20260720.md`](experiments/frontier-split-rootcause-s0-20260720.md)）。三个要点：(1) 「residual 超过 margin」是失败的必要条件但远非充分——good/bad 分裂不能用无方向误差量解释；(2) 23 个 material failure 的 winner-deciding pair 中 21 个落在 TP 轴或 mixed（其余 2 个是 6.2% regret 的边缘 mns-axis case），trace 面的 TP 反序为零；A1 measured collective 把 Qwen235 两个 prefill-only 面的 TP 反序清零（trace-PO p90 regret 21.2%→0.3%）却对 Fixed-PD 完全无效——prefill 路径的 TP 差异化误差源是 collective profile（可修），decode 耦合的 TP 差异化误差是当前所有 material failure 的载体；(3) 「Fixed-PD 失败因为高压」被否证：失败 Fixed-PD 的真机 in-flight（14.05）低于全对的 Trace-PD（38.69），且低压 Fixed-PD 同样失败、失败 objective 随负载切换。另有次要缺陷：14 个 winner 错位来自 simulator 对 MNS 逐位不敏感的精确 tie。
+>
+> **2026-07-20 root-cause update：** Q30 低压 Fixed-PD 的 exact stage ledger 关闭了最后一个未解释的 material failure。相同 batch=1 state 下 Frontier full predictor 给 TP4 `18.3515 ms/step`、TP1 `19.3506 ms/step`，方向正确；但 per-GPU 固定到达率使 cluster arrival 随 TP 增长，叠加 decode residence 高估后，TP4 在 simulator 内自激到 time-weighted batch `3.0437`（96.13% decode 时间 batch≥3），own-state step 变为 `28.1712 ms`。其中相对 batch=1 的 `+9.8197 ms` 有 `+8.9297 ms` 来自 batch-conditioned MoE，collective 仅 `+0.0121 ms`。因此 Fixed-PD 的根因不是“固定 workload”或“高压力”本身，而是 **execution-time residual 进入离散事件时钟后改变 future scheduler state；该 state 再通过 MoE/profile/graph 或 MNS admission 非线性放大，形成 action-dependent signed residual 并穿过 decision margin**。Q30 低压是平滑 state-feedback，Q30 高压是跨 MNS cap 的 threshold amplification，Q235 是 composition drift；三者为同一闭环机制族。
+>
+> **2026-07-20 load-audit update：** Trace-PD overload 不是 Fixed/Trace good-bad 分裂的统一解释。旧 Q30 Trace-PD decode offered/observed-peak throughput≈`1.00×`、peak Running/Waiting=`47/0`；降到 `0.10 req/s/GPU` 后 TTFT `245.95/685.51 → 228.14/835.38 ms`（mean/p90，不出现 tail collapse），TPOT `13.18/15.39 → 7.91/8.90 ms`。旧 Q235 则是 `3.44×` 明确过载、peak=`116/3`；降到 `0.035 req/s/GPU` 后 TTFT `1141.54/2616.69 → 478.14/1347.75 ms`，TPOT `61.89/78.62 → 24.00/28.49 ms`。旧 surface 仍有 `417×/32.6×` mean-TTFT spread，否定“所有配置一样差”。八 case baseline 与 claim boundary 见 [`experiments/frontier-eightcase-load-audit-20260720.md`](experiments/frontier-eightcase-load-audit-20260720.md)。
 
 ## 一眼看懂
 
 - **Topic / problem：** LLM serving 的自动、低成本配置调优（AITuner）。当前主线问题：用 simulator 给部署配置（并行度、批量上限等）排序，什么时候可信？需要补多少真机证据？算上这些成本还划算吗？
 - **Central claim：** simulator 要能帮助配置调优，必须先满足 scheduler transition 的 liveness/coverage，再满足「配置相关残差小于真机 decision margin」；前者决定 capacity 是否有定义，后者决定排序是否正确。(ID: C0)
 - **当前结论：** 早先 35 个 trace stall 不是 Frontier scheduler liveness failure：adapter 为不满 16-token 的 prefix block 错误生成了 cache identity，Frontier 又没有 fail-fast。修正为完整 block、使用真实 graph buckets/KV blocks 和 `piecewise`/`KERNEL_ONLY` profile 后，Qwen30 Trace-PD 的全部 12 个 cell 完成 129/129 request，Frontier 对 TTFT/TPOT/E2E 的 6 个 argmin 均与三次 fresh-server 真机一致；但绝对 latency 仍高估 4--511×。这只证明一个 MoE Trace-PD surface 的 selection fidelity，不能外推到 prefill-only、fixed workload 或 235B。
-- **最大 uncertainty / risk：** trace 面成功的原因未判——是 decode 耦合的 TP 差异化误差在 trace 状态分布下真的变小，还是只是被宽 TP margin 掩盖。这决定「sim 剪枝可信域」的边界怎么画，也决定加压 trace 是否会失败。
-- **下一项 critical action：** 判别两个失败机制假设（离散阈值转换 vs 闭环 state 漂移）：先做零 GPU 的 sim-only jitter 反事实与 Q30 低压 Fixed-PD TPOT 反转分量定位，再按结果决定 dash1–4 上的加压 Trace-PD / jittered Fixed-PD 真机判别面。
+- **最大 uncertainty / risk：** 根因已收敛，且 overload 已被排除为统一解释，但可信域边界仍未画清：trace 面的 heterogeneity 是否让 closed-loop state residual 变小，还是当前 success 主要由 capacity/MNS margin 保护？两个降载点只建立 reference-config latency baseline，不能证明新负载下全 surface 仍选对。
+- **下一项 critical action：** 不再做无锚点的 jitter 猜测；保持 request shape 不变，在预测的 MoE/MNS knee 两侧做小规模 rate sweep，并用少量真机 state/batch anchor 验证 `λR(B)` fixed point。成功标准是同时预测 state-regime、排名与 knee，而不只是某个点的 regret。
 - **停止条件：** T1 出 verdict 且成本账本建立后：pass 且摊销论证成立 → 转向「sim 剪枝 + 真机终选」的 hybrid 机制设计；fail → 转入失败机制归因；两条路都无 insight 增量 → 收敛写作。
 
 ## 核心概念
@@ -39,7 +43,7 @@
     - **Hypothesis（机制，active）：** 误差机制是 action-conditioned residual——执行状态的转移（并行拓扑、kernel family、graph mode、batch 组成）使按算子 profile 的组合预测跨配置不可复合；残差大于 margin 时排序失败。(ID: H2；supported，已细化)
       - **Supporting：** 三个 TP 档的端到端校准系数为 0.72/0.47/0.35，残差确实随配置剧烈变化；235B 的批量上限交互预测错误但被 2× margin 容忍；30B prefill-only 在低负载近似对齐、饱和后按 TP 反向放大，最终 τ-b=−1。
       - **细化（2026-07-20 统一普查）：** 决策相关的残差分量集中在 TP 轴且由 decode 状态耦合产生——prefill-only 面的绝对 scale 仅 0.96–1.37× 且 measured collective 即可清除其 TP 反序，而含 decode 的面 scale 4.3–130×、全部 material failure 都由 TP/mixed pair 决定。「residual>margin」只是必要条件；失败还需要残差对准 winner-deciding pair。
-      - **下一步：** 判别两个 competing 机制假设——离散阈值转换（admission cap/profile cliff 把均匀高估转成差异化误差；打破 fixed workload 的同步性应恢复排序）vs 闭环 state 漂移（打破同步性也救不了）。最便宜路径：sim-only jitter 反事实 + Q30 低压 Fixed-PD TPOT 反转的分量定位（唯一无机制解释的 material failure）。
+      - **机制 verdict（2026-07-20）：** closed-loop state drift 是根因，离散阈值是其放大器而非 competing explanation。Q30 低压 exact ledger 显示同 state 的 TP 方向正确，但 TP4 被模拟 residence 反馈推到 batch 3--4，MoE step 增长后反序；Q30 高压进一步跨过 MNS admission cap；Q235 换成 exact real composition 后排序翻正。下一步从“找根因”转为测量 state-regime/knee 的可信边界。
   - **Subclaim：** 成本论证只有在摊销前提下成立。(ID: C3)
     - **Hypothesis（active）：** 每个 model×硬件×runtime 的一次性对齐成本，摊销到大配置面、频繁重调（引擎版本 churn 的频率证据见 claim map）或禁止在线实验的场景后，低于重复真机调优。(ID: H3；untested——分母已实测，分子未入账)
       - **下一步：** 建 cost ledger（见「下一步」）。
@@ -48,7 +52,7 @@
 
 - **Question：** 生产 trace 忠实回放（prefix 打开、原始到达时间与会话结构）下，best-effort Frontier 能否满足 low-regret gate？
 - **为什么现在做：** 这是 H1 的判决实验；所有已完成的机制分解都在人工 workload 上，不能替代这个 verdict。
-- **当前状态：** Trace-PD 的 graph-aligned surface 已通过 selection gate，但绝对 latency 不通过 calibration。现在以不共享结果的 Fixed-PD、Trace-P、Fixed-P surface 检验它的泛化边界。
+- **当前状态：** Trace-PD 的 graph-aligned surface 已通过原负载 selection gate，但绝对 latency 不通过 calibration；Fixed-PD 的 failure 已定位为 closed-loop state drift。两个降载 Trace-PD anchor 已通过完成率/admission/backlog gate，下一步需要 full surface rate sweep 才能检验 ranking 是否跨 load regime 保持。
 - **Result → decision：** 若其它 surface 排序失败，保留 Trace-PD success 为条件化 envelope，并按 fixed/trace/prefill/decode 的差异定位 state composition；若都通过，才扩大到 Q235 或寻找 simulator 已解决范围之外的新问题。
 - **Experiment card：** [`../runs/frontier-fidelity-envelope-v1/experiment-card.md`](../runs/frontier-fidelity-envelope-v1/experiment-card.md)
 
@@ -56,19 +60,19 @@
 
 - **E1（否证「prefill-only 是充分 easy condition」，支持 H2）：** 30B BF16、去掉 decode/prefix/混合 batch 后，真机最优是 TP4（8 vs 7 req/s/GPU），simulator 却把 TP4 排最差（6 vs 8）：top set 无交集，regret 12.5%，τ-b=−1。产物：`../runs/frontier-phase-factorial-v0/results/final/`（dash0，12.07 H20-GPUh）。
 - **E2（统一机制普查：material failure 全部由 decode 耦合的 TP 差异化误差决定，支持 H2 细化）：** 对 14 个 frozen surface、70 个 case×objective 的方向化分解显示：23 个 material failure 中 21 个由 TP/mixed pair 决定（仅 2 个 6.2% 边缘 mns-axis case）、trace 面 TP 反序为零；measured collective（A1）把 Qwen235 两个 prefill-only 面的 TP 反序清零（trace-PO p90 regret 21.2%→0.3%）但对 Fixed-PD 的 33% 无效；「residual>margin」仅为失败的必要条件。产物：[`../runs/frontier-split-rootcause-v0/results/`](../runs/frontier-split-rootcause-v0/results/decomposition.md)（实验 card：[`experiments/frontier-split-rootcause-s0-20260720.md`](experiments/frontier-split-rootcause-s0-20260720.md)）。
-- **E3（selection 与 calibration 分开）：** 用完整 block projection、`piecewise` 和 graph-compatible KERNEL_ONLY profile 后，Qwen30 Trace-PD 的 12/12 sim cells 完成，6 个 mean/p90 latency objective 的 argmin 都与真机一致；但 sim/real latency ratio 仍为 4--511×。产物：[`../runs/frontier-fidelity-envelope-v1/graph-piecewise-experiment-card.md`](../runs/frontier-fidelity-envelope-v1/graph-piecewise-experiment-card.md)。
+- **E3（closed-loop state 是 Fixed-PD 根因，而非同 state predictor 反序）：** Q30 低压相同 batch=1 state 下 TP4 比 TP1 快约 1.00 ms/step，但 TP4 own state 的 time-weighted batch=3.0437，使 step 增加 9.8197 ms（其中 MoE +8.9297 ms）并反序；Q235 用 exact real composition 重放也把 TP8−TP4 从错向 −20.07 ms 翻为正确 +10.90 ms。Q30 高压再由 MNS cap 将同族 state/residence 误差放大成约 27 s 排队。产物：[`experiments/frontier-split-rootcause-s1-20260720.md`](experiments/frontier-split-rootcause-s1-20260720.md)。
 
 ## 下一步（最多 3 项）
 
-- [ ] **完成 Qwen30 remaining matrix（direct progress）：** Fixed-PD、Trace-P、Fixed-P 分别冻结 Frontier/真机 12-cell surface；P-only 的 TPOT 一律为 N/A。
+- [ ] **画可信域边界：** 固定 request shape，在预测的 MoE/MNS knee 两侧做最小 rate sweep；只在判别点补真机 batch/state anchor，验证 `B≈min(MNS, λR(B))` 是否同时解释 state 与 ranking。
 - [ ] **Q235 portability gate：** 先验证 vLLM0.20 TP4/TP8 FP8 runtime 和 deadc4a profile provenance，再决定是否允许其 Fixed-P sweep。
 - [ ] **建 cost ledger：** parent H3；完成标准 = 每 case 一行（profiling GPU-h、补丁工时、校准探测、sim CPU-h），与已实测的真机调优成本同表，随每个 case 更新。
 
 ## Blocker 或 anomaly
 
-- **下一启动已准备：** dash0 8×H20 当前空闲；graph-compatible attention/linear/MoE/router kernel-only profile 会先在一 GPU smoke，成功后以 3 张 GPU 并行完成 attention shards。完整 replay 仍为 CPU-only。
+- **当前运行状态：** 八 case load audit 的新增真机 run 已完成；未启动 full-surface rate sweep，避免把两个 single-config anchor 外推成 ranking claim。自 2026-07-20 起，本任务只允许使用 `dash1`--`dash4`（每台 8×H20、最多四组并行）；`dash0` 保留给其他同事，不做 probe、同步或运行。
 - **Anomaly（保留）：** 235B pilot 中 simulator 把 10/34 个 anchor 误判为不可行——false-infeasible 是 H1 的主要威胁模式，T1 分析时须单独报告。
-- **勘误（已固定）：** 统一平台为 dash0-only，早期 dash1 出处是文档错误；fixed-shape pilot 的主 SLO（TPOT 40ms）无判别力，150ms 是事后明示的敏感性分析，不得写成盲选的 primary。
+- **平台边界（已更新）：** 历史结果仍来自其各自 card 记录的平台，不改写 provenance；后续实验平台切换为 `dash1`--`dash4`。跨主机比较前必须跑相同 canary 并量化 host effect。fixed-shape pilot 的主 SLO（TPOT 40ms）无判别力，150ms 是事后明示的敏感性分析，不得写成盲选的 primary。
 
 ## Related work