Skip to content

Add DSV4 B200 disaggregated Dynamo SGLang STP configuration / 新增 DSV4 B200 分离式 Dynamo SGLang STP 配置 - #2251

Closed
RohitNagraj wants to merge 6 commits into
mainfrom
dsv4-fp4-b200-dynamo-sglang
Closed

Add DSV4 B200 disaggregated Dynamo SGLang STP configuration / 新增 DSV4 B200 分离式 Dynamo SGLang STP 配置#2251
RohitNagraj wants to merge 6 commits into
mainfrom
dsv4-fp4-b200-dynamo-sglang

Conversation

@RohitNagraj

@RohitNagraj RohitNagraj commented Jul 16, 2026

Copy link
Copy Markdown
Collaborator

Summary

  • Add a B200 multinode disaggregated Dynamo SGLang STP configuration for DSV4 at 8k/1k.
  • Add ten checked-in srt-slurm recipes matching the configured prefill and decode topologies.
  • Resolve the staged DSV4 checkpoint through the B200 launcher and use the checkpoint's native quantization metadata.

Validation

  • Generated all configured sweep combinations successfully.
  • Validated launcher shell syntax, YAML parsing, recipe/config consistency, and changelog ordering.
  • Passed the matrix-logic test suite.

中文说明

  • 新增 DSV4 在 B200 上的多节点分离式 Dynamo SGLang STP 8k/1k 配置。
  • 新增十个已检入的 srt-slurm recipe,与配置中的预填充和解码拓扑一一对应。
  • 通过 B200 启动器解析已暂存的 DSV4 检查点,并使用检查点原生的量化元数据。

验证

  • 已成功生成全部配置扫描组合。
  • 已验证启动器 Shell 语法、YAML 解析、recipe 与配置的一致性以及变更日志顺序。
  • 已通过矩阵逻辑测试套件。

@github-actions

Copy link
Copy Markdown
Contributor

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

@RohitNagraj RohitNagraj changed the title Add DSV4 FP4 B200 Dynamo SGLang configuration Add DSV4 FP4 B200 Dynamo SGLang configuration / 新增 DSV4 FP4 B200 Dynamo SGLang 配置 Jul 16, 2026
Comment thread perf-changelog.yaml
Comment thread perf-changelog.yaml
@github-actions

Copy link
Copy Markdown
Contributor

Ankur-singh added a commit that referenced this pull request Jul 18, 2026
Merge current origin/main and resolve perf-changelog.yaml by retaining every main entry and re-appending PR #2251's entry at the end.\n\n中文:合并最新的 origin/main,并保留主分支的全部 perf-changelog.yaml 条目,将 PR #2251 的专属条目重新追加到文件末尾。
Merge current origin/main and resolve perf-changelog.yaml by retaining every main entry and re-appending PR #2251's entry at the end.

中文:合并最新的 origin/main,并保留主分支的全部 perf-changelog.yaml 条目,将 PR #2251 的专属条目重新追加到文件末尾。
@Ankur-singh
Ankur-singh force-pushed the dsv4-fp4-b200-dynamo-sglang branch from 5a0a86a to 8ac729b Compare July 18, 2026 18:57
@github-actions

Copy link
Copy Markdown
Contributor

@github-actions

Copy link
Copy Markdown
Contributor

1 similar comment
@github-actions

Copy link
Copy Markdown
Contributor

Update the staged checkpoint path and public-facing DSV4 wording while preserving the existing recipe settings.\n\n中文:更新已暂存的 DSV4 检查点路径和面向用户的 DSV4 表述,同时保留现有 recipe 设置。
Resolve the launcher integration against the current B200 runner and rebuild perf-changelog.yaml from main with the DSV4 entry appended last.\n\n中文:基于当前 B200 启动器解决集成冲突,并以 main 为基准重建 perf-changelog.yaml,将 DSV4 条目追加到文件末尾。
@RohitNagraj RohitNagraj changed the title Add DSV4 FP4 B200 Dynamo SGLang configuration / 新增 DSV4 FP4 B200 Dynamo SGLang 配置 Add DSV4 B200 disaggregated Dynamo SGLang STP configuration / 新增 DSV4 B200 分离式 Dynamo SGLang STP 配置 Jul 31, 2026
@github-actions

github-actions Bot commented Aug 2, 2026

Copy link
Copy Markdown
Contributor

@RohitNagraj

Copy link
Copy Markdown
Collaborator Author

Moved to #2560 .

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

Development

Successfully merging this pull request may close these issues.

2 participants