[AgentX]: Use burst phase starts by default / [AgentX]：默认使用阶段起始突发模式

[cquil11]

Summary

• Bump the utils/aiperf submodule from f5ae8134 to 208125ac, the merge commit on cquil11/aiperf-agentx-v1.0.
• Make Agentic Replay burst warmup and profiling phase starts by default in AIPerf, so configured concurrency becomes active within the benchmark window.
• Preserve recorded inter-turn timing; setting burst_phase_starts: false restores spread phase starts.
• Keep the behavior in AIPerf itself—there is no duplicate InferenceX wrapper or new InferenceX test file.

Merged AIPerf change: SemiAnalysisAI/aiperf#22

Discovered while bringing up Kimi K3 GB200 AgentX in #2354. Recorded initial offsets reached 56,048.7 seconds at concurrency 32 and 252,825.1 seconds at concurrency 256, so spread phase starts could prevent the requested concurrency from becoming active during the run.

Validation

• AIPerf targeted suite: 758 passed
• AIPerf pre-commit: passed
• InferenceX matrix tests: 224 passed

中文说明

• 将 utils/aiperf 子模块从 f5ae8134 更新到 cquil11/aiperf-agentx-v1.0 分支上的合并提交 208125ac。
• 在 AIPerf 中将 Agentic Replay 的预热和性能测试阶段起始请求默认改为突发发送，确保配置的并发度能在基准测试窗口内实际生效。
• 保留轨迹中的请求间时序；设置 burst_phase_starts: false 可恢复分散式阶段启动。
• 该行为直接在 AIPerf 中实现；InferenceX 不再包含重复的封装逻辑，也不新增 InferenceX 测试文件。

已合并的 AIPerf 变更：SemiAnalysisAI/aiperf#22

该问题是在 #2354 的 Kimi K3 GB200 AgentX 启动调试中发现的。并发度为 32 和 256 时，记录的初始偏移分别达到 56,048.7 秒和 252,825.1 秒，分散式阶段启动可能导致目标并发度在测试期间始终无法生效。

验证

• AIPerf 定向测试：758 项通过
• AIPerf pre-commit：通过
• InferenceX matrix 测试：224 项通过

[github-actions]

Thanks for the contribution! Please reach out to respective companies' CODEOWNER to fill in the latest PR_REVIEW_CHECKLIST.md before pinging core maintainer on Slack for review. In order for the signoff PR check bot to trigger, you must follow the PR_REVIEW_CHECKLIST.md template correctly, including the phrase As a PR reviewer and CODEOWNER, I have reviewed this and have.

For PR verification, add the full-sweep-fail-fast label (strongly recommended) to this PR — the benchmark sweep only runs on labeled PRs. Use full-sweep-enabled only if you need matrix jobs to keep running past a failure.

PR authors are responsible for ensuring that after merging, all GitHub Action jobs fully pass. A lot of the time, failures are just flakes and simply re-running the failed jobs will fix it. See GitHub's docs on re-running failed jobs

---

感谢你的贡献！请联系相应公司的 CODEOWNER 填写最新的 PR_REVIEW_CHECKLIST.md，然后再在 Slack 上联系核心维护者进行审阅。为了触发 signoff PR 检查机器人，你必须正确遵循 PR_REVIEW_CHECKLIST.md 模板，包括保留英文语句 As a PR reviewer and CODEOWNER, I have reviewed this and have。

如需进行 PR 验证，请为此 PR 添加 full-sweep-fail-fast 标签（强烈推荐）— 基准测试 sweep 仅在带有标签的 PR 上运行。仅当需要矩阵任务在失败后继续运行时才使用 full-sweep-enabled。

PR 作者有责任确保合并后所有 GitHub Action 任务完全通过。 很多时候失败只是偶发抖动（flake），重新运行失败的任务即可解决。参见 GitHub 关于重新运行失败任务的文档

[claude]

Claude finished @cquil11's task in 2m 26s —— View job

---

PR Review

• Gather context and read the PR diff
• Verify the submodule pointer change
• Check for stale InferenceX-side references
• Post review

LGTM - no blocking issues found.

The diff is exactly the one-line utils/aiperf submodule bump described (f5ae8134 → 208125ac). Verified on the InferenceX side:

• No stale references to burst_phase_starts / AIPERF_BURST_PHASE_STARTS remain anywhere in InferenceX, consistent with the decision to keep the behavior in AIPerf and drop the wrapper plumbing.
• .gitmodules already tracks cquil11/aiperf-agentx-v1.0, matching the branch the pinned merge commit lives on — no config drift introduced.
• No master configs or benchmark scripts touched, so no perf-changelog entry is required.

Note: I couldn't inspect the AIPerf commit contents directly (submodule not initialized in this environment and network access to the aiperf repo is restricted), so the behavioral review of the default change itself rests on the merged upstream review in SemiAnalysisAI/aiperf#22 and the reported validation (758 AIPerf tests, 224 matrix tests).

[claude]

LGTM - this is a one-line submodule pointer bump (utils/aiperf to 208125ac) pulling in an already-merged, reviewed upstream change that flips burst_phase_starts to default True. The one finding is a doc-only nit (stale docstring text) inside the submodules own commit history with no functional impact.

Extended reasoning...

Overview

The entire InferenceX-side diff is a single line: the utils/aiperf submodule pointer moves from f5ae8134 to 208125ac, the merge commit for cquil11/aiperf-agentx-v1.0 (upstream PR SemiAnalysisAI/aiperf#22). There is no InferenceX wrapper code, config, or test file touched - the behavioral change (Agentic Replay warmup/profiling phase starts default to burst instead of spread) lives entirely inside the aiperf submodule.

Security risks

None. This is a benchmarking-tool timing/dispatch behavior change with no auth, crypto, injection, or data-exposure surface.

Level of scrutiny

Low-to-moderate is appropriate: the InferenceX-visible change is a mechanical submodule bump, so I read into the submodule itself to verify the underlying commit (fb39d0df) actually does what the PR description claims. I confirmed the config default flip in loadgen_config.py (False to True), the corresponding docstring update in docs/cli-options.md, and that the dispatch logic in agentic_replay.py (spread = not self._burst_phase_starts, and the burst/spread branches in _execute_warmup and _dispatch_snapshot_for_profiling) correctly implements the new default - no functional bug there. The behavior is escapable via burst_phase_starts: false for anyone who needs the old faithful-replay timing.

Other factors

The one flagged issue (stale spread-is-default wording left in two docstrings/comments in the same file, contradicting the now-accurate module-level docstring) is cosmetic, pre-existing in the upstream commit being pulled in, and does not affect runtime behavior - the actual branch conditions were updated correctly everywhere checked. It does not warrant blocking this bump; a follow-up doc fix in the upstream aiperf repo is sufficient. The PR description also documents concrete motivation (concurrency never reaching target under spread mode, evidenced by large recorded offsets at concurrency 32/256) and reports both AIPerf's and InferenceX's test suites passing.

[claude]

🟡 The aiperf submodule bump (fb39d0df) flips burst_phase_starts to default True and updates the module-level docstring in agentic_replay.py accordingly, but two other docstrings in the same file — _execute_warmup's "Dispatch timing" docstring (~line 496) and the inline comment in _dispatch_snapshot_for_profiling (~line 1499) — still describe spread as the default and burst as the opt-in flag, now contradicting actual behavior. This is a pre-existing issue in the upstream aiperf commit being pulled in, doc-only with no functional impact, and would need a follow-up fix + re-bump in the aiperf repo.

Extended reasoning...

What the bug is: Commit fb39d0df ("refactor(agentic): burst phase starts by default") in the bumped utils/aiperf submodule flips burst_phase_starts from False to True (see loadgen_config.py:278) and correctly updates the module-level docstring and the __init__ comment in src/aiperf/timing/strategies/agentic_replay.py to describe burst as the new default. However, two other docstrings in the same file that describe the same flag were left untouched and now say the opposite of what's true.

Where it manifests:

1. _execute_warmup's "Dispatch timing" docstring (~line 496-507) still reads: Default (spread): aligned GLOBALLY... and frames --burst-phase-starts as the override that changes behavior away from the default.
2. The inline comment inside _dispatch_snapshot_for_profiling (~line 1499-1501) still reads: Spread (default): t0 = 0 ... Burst (--burst-phase-starts): t0 = the lane's min offset.

Both explicitly label "spread" as the default, which was true before this commit but is now false — the default is burst.

Why the existing update missed these: The diff for fb39d0df only touched the module docstring (lines ~20-40) and the __init__ comment (lines ~215-221), which are the two most visible places describing the flag. It missed that the same "default vs. override" framing was duplicated verbatim in two method-level docstrings further down in the file, so those got left behind when the semantics flipped.

Impact: This is purely a documentation/comment inconsistency — the actual dispatch logic in both methods is correct and unaffected (the code paths still correctly branch on self._burst_phase_starts). No runtime behavior changes. The only impact is that a future reader of _execute_warmup or _dispatch_snapshot_for_profiling in isolation would be misled about which behavior is the current default, since these method-level docs now contradict the accurate module-level docstring in the same file.

Proof (step-by-step):

1. Before fb39d0df: burst_phase_starts defaulted to False (spread). Module docstring, __init__ comment, _execute_warmup docstring, and _dispatch_snapshot_for_profiling comment all consistently said "spread is default."
2. fb39d0df changes loadgen_config.py's default to True (burst) and updates the module docstring (~line 24: "By default the priming requests BURST together...") and __init__ comment (~lines 215-216: "Real loadgen configs default to burst...").
3. The commit's diff (visible in the verifier logs) shows no changes to the Default (spread): string in _execute_warmup or the Spread (default): string in _dispatch_snapshot_for_profiling.
4. Reading those two methods today: _execute_warmup says spread is default while the module docstring 60 lines above it says burst is default — a direct, verifiable contradiction within the same file.

How to fix: In the upstream aiperf repo, swap the "default"/"override" labeling in both the _execute_warmup docstring and the _dispatch_snapshot_for_profiling inline comment to match the new burst-by-default semantics (mirroring the wording already used in the module docstring), then re-bump the utils/aiperf submodule pointer in InferenceX. This doesn't need to block the current PR since it's doc-only and pre-existing in the commit being pulled in — a quick follow-up is sufficient.