Killing `study tune` with a default SIGTERM skipped the finally blocks, leaving the
vLLM engine and its EngineCore workers (which inherit the AITUNER_* marker env) alive
on the GPUs — twice leaking GPU memory that needed a root reset. Install a SIGTERM
handler in run_trial that raises KeyboardInterrupt so _terminate_process_tree runs,
ignore SIGTERM during teardown so a second signal can't re-orphan it, and restore the
prior handler afterward. Main-thread-guarded; unit-tested.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
stream_chat_completion (and the LLM stream/chat paths) only caught HTTPError, so a
request exceeding request_timeout_s raised a raw TimeoutError mid-stream that escaped
_run_one_request (which only catches HttpClientError), propagated through the probe,
and crashed the whole trial ("failed: timed out"). A timed-out request is a failed
request (SLO miss), not a trial crash. Catch OSError (covers TimeoutError, URLError,
ConnectionError) after HTTPError and wrap it. Exposed by lowering request_timeout_s
to 180s on the 27B run.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
When a truncated probe's measured pass-rate lands within trace.adaptive_stop.
boundary_delta of the SLO target, re-measure on the full window and use that
verdict. Offered-L-C-A convergence cannot see engine-state drift in the window
tail, so a near-knee truncated verdict is untrustworthy (validated: prefix 0.96
vs full 0.946 at threshold 0.08594). The guard fires only on feasibility-knee
probes, so non-boundary probes keep the Stop-A saving. Default delta=0.02.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 4 of the two-stop work. The harness already pre-empts the LLM with
deterministic stops and guided probes, but an LLM-originated should_stop could
still end the loop while the validator saw remaining opportunity.
Add harness._stop_authority, exposed as context["stop_authority"], whose
`authorized` mirrors the deterministic harness stop decision and whose
`opportunity_remains` flags an open topology frontier or a high-value planned
candidate. In study tune, an LLM-originated should_stop is now honored only when
the validator authorizes it; an unauthorized stop is vetoed (bounded budget) so
the loop cannot converge prematurely on the agent's say-so. File- and
harness-originated stops are unaffected, and the stop reason chain is recorded.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 2 of the two-stop work. The L-C-A vector is a deterministic function of the
trace's offered metadata, so the convergence of prefix-vs-full L-C-A (the paper's
Fig. 9 curve) can be computed up front rather than monitored live, with identical
result and no per-request overhead.
- lca.find_convergence_prefix: earliest arrival-ordered prefix whose L and A family
similarities reach tau and the slow C family reaches the stricter tau_c for
stable_checks consecutive checkpoints. Self-similarity uses the raw log-feature
vector (same window -> identical per-dim spread; RobustScaler is reserved for the
cross-window Stop-C). If C never converges it reports the full set, which is the
C-gate: no early stop on a cold/under-warmed cache. The checkpoint sims double as
Phase 3 calibration data.
- spec.AdaptiveStopSpec (trace.adaptive_stop), disabled by default until the
thresholds are calibrated, so existing studies are unaffected.
- worker._adaptive_replay_set truncates each probe's replay to the convergence
prefix and records a certificate (converged, fraction, family similarity) into
probe history and probe_details. Offered request_rate at the threshold is
unchanged; only wall-clock replay shrinks.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 1 of the two-stop work. Subsampling the trace by per-request uniform score
broke multi-turn sessions (a kept turn-2 could lose its turn-1), which lowered the
realized KV-cache hit rate as offered load dropped — so the feasibility boundary
was measured on a workload with a different C than production, contradicting the
paper's scale-stationary L-C-A premise.
prepare_trace_windows now resolves each row's session root via the parent_chat_id
chain in a single streaming pass and assigns sampling_u per session, so thresholding
keeps or drops whole sessions and preserves intra-session prefix reuse. Rows whose
parent fell outside the span fall back to grouping under the parent id.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 0 of the two-stop work. The prompt block labeled `workload_lca_profile`
previously re-derived L-C-A from summarize_window's ad-hoc percentiles, diverging
from the paper's 10-dim RobustScaler vector implemented in lca.py. Make that block
authoritative: build_harness_context now accepts an optional workload_profile and
renders the canonical 10-dim vector + per-family stats when present, falling back
to the legacy rendering only when no profile is supplied (direct unit-test calls).
Real call sites (study prompt/llm-propose/tune, run_baseline_then_llm) build the
profile via lca.build_study_workload_profile and pass it through build_prompt. The
heuristic regime classifiers keep reading window_summary; that is the heuristic
layer, distinct from the similarity metric.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Implement the paper's 10-dimensional L-C-A workload feature vector
(RobustScaler-normalized, sim=exp(-||dz||)) in lca.py, and wire it into
`aituner profile window` / `aituner profile similarity`. Covered by tests.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
