[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78D17D18CKA3HGZC9MXH0RF":3,"skill-same:01M78D17D18CKA3HGZC9MXH0RF":72},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":14,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M78D17D18CKA3HGZC9MXH0RF","nemo-mbridge-perf-tp-dp-comm-overlap",null,"Operational guide for enabling TP, DP, and PP communication overlap in Megatron-Bridge, including config knobs, code anchors, pitfalls, and verification.","Solizardking\u002Fskillhub-main","Solizardking","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F108096810?v=4","01M786KDB10AYV042N0F9BD5B5","Clawd","01M786NYYXSZK1D05WM8JQPC48",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fnvidia\u002Fnemo-mbridge-perf-tp-dp-comm-overlap\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: nemo-mbridge-perf-tp-dp-comm-overlap\ndescription: Operational guide for enabling TP, DP, and PP communication overlap in Megatron-Bridge, including config knobs, code anchors, pitfalls, and verification.\nlicense: Apache-2.0\nwhen_to_use: Enabling TP\u002FDP\u002FPP comm overlap, or tracing a throughput regression to a comm overlap config change; 'overlap_param_gather', 'overlap_grad_reduce', 'sequence-parallel overlap', 'TP overlap', 'DP overlap', 'comm overlap'.\n---\n\n# TP \u002F DP \u002F PP Communication Overlap Skill\n\nFor stable background and recommendation level, see:\n\n- @docs\u002Ftraining\u002Fcommunication-overlap.md\n\n## Enablement\n\nMinimal Bridge override:\n\n```python\nfrom megatron.bridge.training.comm_overlap import CommOverlapConfig\n\ncfg.model.tensor_model_parallel_size = 4\ncfg.model.sequence_parallel = True\ncfg.model.pipeline_model_parallel_size = 4\ncfg.model.virtual_pipeline_model_parallel_size = 2\n\ncfg.comm_overlap = CommOverlapConfig(\n    tp_comm_overlap=True,\n)\n\ncfg.ddp.use_distributed_optimizer = True\ncfg.ddp.overlap_grad_reduce = True\ncfg.ddp.overlap_param_gather = True\n```\n\nOptional TP preset:\n\n```python\nfrom megatron.bridge.training.comm_overlap import userbuffers_bf16_h100_h12288_tp4_mbs1_seqlen2048\n\ncfg.comm_overlap.tp_comm_overlap_cfg = userbuffers_bf16_h100_h12288_tp4_mbs1_seqlen2048\n```\n\nPrecision knobs belong to mixed precision:\n\n```python\ncfg.mixed_precision.grad_reduce_in_fp32 = False\ncfg.mixed_precision.fp8_param_gather = False\n```\n\n## Code Anchors\n\nBridge overlap gating:\n\n```439:449:src\u002Fmegatron\u002Fbridge\u002Ftraining\u002Fcomm_overlap.py\nif self.user_comm_overlap_cfg.tp_comm_overlap is True:\n    if model_cfg.tensor_model_parallel_size \u003C 2:\n        ...\n    elif not model_cfg.sequence_parallel:\n        ...\n    elif not HAVE_TE:\n        ...\n```\n\nPP overlap selection:\n\n```451:458:src\u002Fmegatron\u002Fbridge\u002Ftraining\u002Fcomm_overlap.py\nif model_cfg.pipeline_model_parallel_size > 1:\n    if vp_size > 1:\n        comm_overlap_cfg.overlap_p2p_comm = True\n        comm_overlap_cfg.batch_p2p_comm = False\n    else:\n        comm_overlap_cfg.overlap_p2p_comm = False\n        comm_overlap_cfg.batch_p2p_comm = True\n```\n\nDP overlap defaults:\n\n```572:579:src\u002Fmegatron\u002Fbridge\u002Ftraining\u002Fcomm_overlap.py\nif self.data_parallel_size > 1:\n    comm_overlap_cfg.bucket_size = 128 * 1024 * 1024\n    comm_overlap_cfg.overlap_grad_reduce = True\n    comm_overlap_cfg.overlap_param_gather = True\n```\n\nLaunch-time env tuning:\n\n```570:609:src\u002Fmegatron\u002Fbridge\u002Frecipes\u002Frun_plugins.py\nexecutor.env_vars[\"CUDA_DEVICE_MAX_CONNECTIONS\"] = str(cuda_device_max_connections)\n...\nexecutor.env_vars[\"NVTE_FWD_LAYERNORM_SM_MARGIN\"] = str(self.layernorm_sm_margin)\nexecutor.env_vars[\"NVTE_BWD_LAYERNORM_SM_MARGIN\"] = str(self.layernorm_sm_margin)\n```\n\n## Pitfalls\n\n1. TP overlap silently disables itself if `sequence_parallel=False` or Transformer Engine is unavailable.\n2. PP overlap is not enabled for all PP cases. Bridge only auto-selects `overlap_p2p_comm=True` when `PP > 1` and `VPP > 1`.\n3. `bucket_size` is a parameter-count knob, not a byte-size knob.\n4. `grad_reduce_in_fp32` and `fp8_param_gather` should be set through mixed precision, not as standalone DDP tuning first.\n5. `CUDA_DEVICE_MAX_CONNECTIONS` and LayerNorm SM margin are launch-time plugin settings, not `CommOverlapConfig` fields.\n\n## Verification\n\nUse the checked-in overlap unit coverage first:\n\n```bash\nuv run python -m pytest tests\u002Funit_tests\u002Ftraining\u002Ftest_comm_overlap.py -q\n```\n\nOptional second check if `nemo_run` is available:\n\n```bash\nuv run python -m pytest tests\u002Funit_tests\u002Frecipes\u002Ftest_run_plugins.py -q\n```\n\nSuccess criteria:\n\n- first command reports `26 passed`\n- second command validates plugin-owned env wiring when not skipped\n","# TP \u002F DP \u002F PP Communication Overlap Skill\n\nFor stable background and recommendation level, see:\n\n- @docs\u002Ftraining\u002Fcommunication-overlap.md\n\n## Enablement\n\nMinimal Bridge override:\n\n```python\nfrom megatron.bridge.training.comm_overlap import CommOverlapConfig\n\ncfg.model.tensor_model_parallel_size = 4\ncfg.model.sequence_parallel = True\ncfg.model.pipeline_model_parallel_size = 4\ncfg.model.virtual_pipeline_model_parallel_size = 2\n\ncfg.comm_overlap = CommOverlapConfig(\n    tp_comm_overlap=True,\n)\n\ncfg.ddp.use_distributed_optimizer = True\ncfg.ddp.overlap_grad_reduce = True\ncfg.ddp.overlap_param_gather = True\n```\n\nOptional TP preset:\n\n```python\nfrom megatron.bridge.training.comm_overlap import userbuffers_bf16_h100_h12288_tp4_mbs1_seqlen2048\n\ncfg.comm_overlap.tp_comm_overlap_cfg = userbuffers_bf16_h100_h12288_tp4_mbs1_seqlen2048\n```\n\nPrecision knobs belong to mixed precision:\n\n```python\ncfg.mixed_precision.grad_reduce_in_fp32 = False\ncfg.mixed_precision.fp8_param_gather = False\n```\n\n## Code Anchors\n\nBridge overlap gating:\n\n```439:449:src\u002Fmegatron\u002Fbridge\u002Ftraining\u002Fcomm_overlap.py\nif self.user_comm_overlap_cfg.tp_comm_overlap is True:\n    if model_cfg.tensor_model_parallel_size \u003C 2:\n        ...\n    elif not model_cfg.sequence_parallel:\n        ...\n    elif not HAVE_TE:\n        ...\n```\n\nPP overlap selection:\n\n```451:458:src\u002Fmegatron\u002Fbridge\u002Ftraining\u002Fcomm_overlap.py\nif model_cfg.pipeline_model_parallel_size > 1:\n    if vp_size > 1:\n        comm_overlap_cfg.overlap_p2p_comm = True\n        comm_overlap_cfg.batch_p2p_comm = False\n    else:\n        comm_overlap_cfg.overlap_p2p_comm = False\n        comm_overlap_cfg.batch_p2p_comm = True\n```\n\nDP overlap defaults:\n\n```572:579:src\u002Fmegatron\u002Fbridge\u002Ftraining\u002Fcomm_overlap.py\nif self.data_parallel_size > 1:\n    comm_overlap_cfg.bucket_size = 128 * 1024 * 1024\n    comm_overlap_cfg.overlap_grad_reduce = True\n    comm_overlap_cfg.overlap_param_gather = True\n```\n\nLaunch-time env tuning:\n\n```570:609:src\u002Fmegatron\u002Fbridge\u002Frecipes\u002Frun_plugins.py\nexecutor.env_vars[\"CUDA_DEVICE_MAX_CONNECTIONS\"] = str(cuda_device_max_connections)\n...\nexecutor.env_vars[\"NVTE_FWD_LAYERNORM_SM_MARGIN\"] = str(self.layernorm_sm_margin)\nexecutor.env_vars[\"NVTE_BWD_LAYERNORM_SM_MARGIN\"] = str(self.layernorm_sm_margin)\n```\n\n## Pitfalls\n\n1. TP overlap silently disables itself if `sequence_parallel=False` or Transformer Engine is unavailable.\n2. PP overlap is not enabled for all PP cases. Bridge only auto-selects `overlap_p2p_comm=True` when `PP > 1` and `VPP > 1`.\n3. `bucket_size` is a parameter-count knob, not a byte-size knob.\n4. `grad_reduce_in_fp32` and `fp8_param_gather` should be set through mixed precision, not as standalone DDP tuning first.\n5. `CUDA_DEVICE_MAX_CONNECTIONS` and LayerNorm SM margin are launch-time plugin settings, not `CommOverlapConfig` fields.\n\n## Verification\n\nUse the checked-in overlap unit coverage first:\n\n```bash\nuv run python -m pytest tests\u002Funit_tests\u002Ftraining\u002Ftest_comm_overlap.py -q\n```\n\nOptional second check if `nemo_run` is available:\n\n```bash\nuv run python -m pytest tests\u002Funit_tests\u002Frecipes\u002Ftest_run_plugins.py -q\n```\n\nSuccess criteria:\n\n- first command reports `26 passed`\n- second command validates plugin-owned env wiring when not skipped\n",{"name":5,"license":15,"description":7,"when_to_use":24},"Enabling TP\u002FDP\u002FPP comm overlap, or tracing a throughput regression to a comm overlap config change; 'overlap_param_gather', 'overlap_grad_reduce', 'sequence-parallel overlap', 'TP overlap', 'DP overlap', 'comm overlap'.","main","be3de0244dc1a57e323d8160787ebb40fb64930e","skills\u002Fnvidia\u002Fnemo-mbridge-perf-tp-dp-comm-overlap","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main","https:\u002F\u002Fskills.x402.wtf","https:\u002F\u002Fgithub.com\u002FSolizardking","https:\u002F\u002Fmusebook.trade","Got a mile high and never came down","2026-10-08 23:20:13",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add Solizardking\u002Fskillhub-main --skill nemo-mbridge-perf-tp-dp-comm-overlap",{"tool":40,"command":41},"bunx","bunx skills add Solizardking\u002Fskillhub-main --skill nemo-mbridge-perf-tp-dp-comm-overlap",{"tool":43,"command":44},"pnpm","pnpm dlx skills add Solizardking\u002Fskillhub-main --skill nemo-mbridge-perf-tp-dp-comm-overlap","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fnvidia\u002Fnemo-mbridge-perf-tp-dp-comm-overlap\u002FSKILL.md",true,[48,54,58,62,66],{"path":49,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":52,"sourceHtmlUrl":53},"BENCHMARK.md","REFERENCE","参考文档",4371,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fnvidia\u002Fnemo-mbridge-perf-tp-dp-comm-overlap\u002FBENCHMARK.md",{"path":55,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":56,"sourceHtmlUrl":57},"card.yaml",2103,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fnvidia\u002Fnemo-mbridge-perf-tp-dp-comm-overlap\u002Fcard.yaml",{"path":59,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":60,"sourceHtmlUrl":61},"evals\u002Fevals.json",1439,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fnvidia\u002Fnemo-mbridge-perf-tp-dp-comm-overlap\u002Fevals\u002Fevals.json",{"path":63,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":64,"sourceHtmlUrl":65},"skill-card.md",3817,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fnvidia\u002Fnemo-mbridge-perf-tp-dp-comm-overlap\u002Fskill-card.md",{"path":67,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":70,"sourceHtmlUrl":71},"skill.oms.sig","OTHER","其他",4833,"https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fnvidia\u002Fnemo-mbridge-perf-tp-dp-comm-overlap\u002Fskill.oms.sig",[73,81,88,95,102,110],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M78E0TGCAMNPS207BA3G07S7","zkrouter","Self-hosted OpenAI-compatible LLM router — smart tier-based routing, ZK-stamped routing decisions, mode overrides, free OpenRouter at install via the Birth bot. Save 60-80% on inference costs by routing to the cheapest capable model across 12+ OpenRouter models.","MIT","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fzkrouter\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M78E0R1Z8BDT6EDTHY17MTNB","zk","For custom ZK Solana programs and privacy-preserving applications to prevent double spending. Guide to integrate rent-free nullifier PDAs for double-spend prevention.","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fzk\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":19},"01M78E0P3ZTZ1YWF8MD5MM23QE","zk-omni-messaging","Zero-knowledge omnichain messaging between Robinhood Chain and Solana using CheshireZkOmniMessenger (msgType 4) with nullifier anti-replay and the zk-omni-relayer service.","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fzk-omni-messaging\u002FSKILL.md",[],[],{"skillNo":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":19},"01M78E0JXZ2PNB1896VXAGDBWD","zerion-wallet-data","Use when the user asks to check a wallet's portfolio, token balances, DeFi positions, PnL, transaction history, or NFTs. Provides interpreted on-chain data across 40+ EVM chains and Solana via x402 pay-per-request or API key — no raw RPC decoding needed.","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fzerion-wallet-data\u002FSKILL.md",[],[],{"skillNo":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":106,"sourceHtmlUrl":107,"categories":108,"keywords":109,"publishedAt":19},"01M78E0G14FH9GRRTQYMKD5E59","youtube-clipper","YouTube 视频智能剪辑工具。下载视频和字幕，AI 分析生成精细章节（几分钟级别）， 用户选择片段后自动剪辑、翻译字幕为中英双语、烧录字幕到视频，并生成总结文案。 使用场景：当用户需要剪辑 YouTube 视频、生成短视频片段、制作双语字幕版本时。 关键词：视频剪辑、YouTube、字幕翻译、双语字幕、视频下载、clip video","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fyoutube-clipper\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fyoutube-clipper\u002FSKILL.md",[],[],{"skillNo":111,"name":112,"nameZh":6,"description":113,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":77,"licenseFileUrl":6,"sourceHtmlUrl":114,"categories":115,"keywords":116,"publishedAt":19},"01M78E0C2ZYA99CFFDNB7ZMK5S","yield-optimization","Multi-chain yield optimization via yield.xyz (StakeKit). Discover 2,988+ yield opportunities across 75+ blockchains, build deposit\u002Fwithdrawal transactions with shell scripts, and sign them with a MoonPay wallet. Use when the user wants to earn yield, find the best lending or staking rate, enter\u002Fexit a position, or check portfolio balances.","https:\u002F\u002Fgithub.com\u002FSolizardking\u002Fskillhub-main\u002Fblob\u002Fmain\u002Fskills\u002Fyield-optimization\u002FSKILL.md",[],[]]