NVIDIA/cloudai

آخر نشاط مصدر مسجل: ٢٦ أكتوبر ٢٠٢٦ في ١٢:٢٩

آخر تحديث لفهرس SkillsMP: ٥ نوفمبر ٢٠٢٦ في ١٩:٥٧

skills مجمعة: ٦

نجوم GitHub: ١٠٢
تفرعات GitHub: ٦٦

التثبيت بمساعدة مساعد ذكي

انسخ هذا الطلب إلى المساعد الذكي الذي تستخدمه.

Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/NVIDIA/cloudai.

ملف README للمستودع

مصدر ملف README للمستودع: README.md

قراءة الملف الأصلي

عرض ملف README كاملًا

# CloudAI Benchmark Framework CloudAI benchmark framework aims to develop an industry standard benchmark focused on grading Data Center (DC) scale AI systems in the Cloud. The primary motivation is to provide automated benchmarking on various systems. ## Get Started Using `uv` tool allows users to run CloudAI without manually managing required Python versions and dependencies. ```bash git clone git@github.com:NVIDIA/cloudai.git cd cloudai uv run cloudai --help ``` For details on setting up workloads' requirements, please refer to the [installation guide](https://nvidia.github.io/cloudai/workloads_requirements_installation.html) For details and `pip`-based installation, please refer to the [documentation](https://nvidia.github.io/cloudai/#get-started). ## Key Concepts CloudAI operates on three main schemas: - **System Schema**: Describes the system, including the scheduler type, node list, and global environment variables - **Test Schema**: An instance of a test template with custom arguments and environment variables - **Test Scenario Schema**: A set of tests with dependencies and additional descriptions about the test scenario These schemas enable CloudAI to be flexible and compatible with different systems and configurations. ## Support Matrix |Test|Slurm|Kubernetes|RunAI|Standalone| |---|---|---|---|---| |AI Dynamo|✅|✅|❌|❌| |BashCmd|✅|❌|❌|❌| |ChakraReplay|✅|❌|❌|❌| |DDLB|✅|❌|❌|❌| |DeepEP|✅|❌|❌|❌| |DynamoMocker|❌|❌|❌|✅| |Fio|✅|❌|❌|✅| |JaxToolbox workloads (DEPRECATED)|✅|❌|❌|❌| |MegatronRun|✅|❌|❌|❌| |NCCL|✅|✅|✅|❌| |NeMo v1.0 aka NemoLauncher (DEPRECATED)|✅|❌|❌|❌| |NeMo v2.0 (aka NemoRun)|✅|❌|❌|❌| |NIXL benchmark|✅|❌|❌|❌| |NIXL kvbench|✅|❌|❌|❌| |NIXL CTPerf|✅|❌|❌|❌| |SGLang|✅|❌|❌|❌| |Sleep|✅|✅|❌|✅| |SlurmContainer|✅|❌|❌|❌| |Triton Inference|✅|❌|❌|❌| |UCC|✅|❌|❌|❌| |vLLM|✅|❌|❌|❌| Note: Deprecated means that a workload support exists, but we are not maintaining it actively anymore and newer configurations might not work. For more detailed information, please refer to the [official documentation](https://nvidia.github.io/cloudai/workloads/index.html). ## CloudAI Modes Usage Examples ### run This mode runs workloads. It automatically installs prerequisites if they are not met. ```bash cloudai run\ --system-config conf/common/system/example_slurm_cluster.toml\ --tests-dir conf/common/test\ --test-scenario conf/common/test_scenario/sleep.toml ``` ### dry-run This mode simulates running experiments without actually executing them. This is useful for verifying configurations and testing experiment setups. ```bash cloudai dry-run\ --system-config conf/common/system/example_slurm_cluster.toml\ --tests-dir conf/common/test\ --test-scenario conf/common/test_scenario/sleep.toml ``` ### generate-report This mode generates reports under the scenario directory. It automatically runs as part of the `run` mode after experiments are completed. ```bash cloudai generate-report\ --system-config conf/common/system/example_slurm_cluster.toml\ --tests-dir conf/common/test\ --test-scenario conf/common/test_scenario/sleep.toml\ --result-dir /path/to/result_directory ``` Generated artifacts depend on the scenario contents: - A plain scenario status report is written as `<scenario>.html`. - If the scenario contains DSE test cases, an additional DSE-specific report is written as `<scenario>-dse-report.html`. - For DSE runs, the best discovered test configuration is also written as `<dse-case>/<iteration>/<dse-case>.toml`. - Custom reporters could generate additional artifacts. ### install This mode installs test prerequisites. For more details, please refer to the [installation guide](https://nvidia.github.io/cloudai/workloads_requirements_installation.html). It automatically runs as part of the `run` mode if prerequisites are not met. ```bash cloudai install\ --system-config conf/common/system/example_slurm_cluster.toml\ --tests-dir conf/common/test\ --test-scenario conf/common/test_scenario/sleep.toml ``` ### uninstall The opposite to the install mode, this mode removes installed test prerequisites. ```bash cloudai uninstall\ --system-config conf/common/system/example_slurm_cluster.toml\ --tests-dir conf/common/test\ --test-scenario conf/common/test_scenario/sleep.toml ``` ### list This mode lists internal components available within CloudAI. ```bash cloudai list <component_type> ``` ### verify-configs This mode verifies the correctness of system, test and test scenario configuration files. ```bash # verify all at once cloudai verify-configs conf # verify a single file cloudai verify-configs conf/common/system/example_slurm_cluster.toml # verify all scenarios using specific folder with Test TOMLs cloudai verify-configs --tests-dir conf/release/spcx/l40s/test conf/release/spcx/l40s/test_scenario ``` ## Additional Documentation For more detailed instructions and guidance, including advanced usage and troubleshooting, please refer to the [official documentation](https://nvidia.github.io/cloudai/). ## Contribution Please feel free to contribute to the CloudAI project and share your insights. Your contributions are highly appreciated. ## License This project is licensed under Apache 2.0. See the LICENSE file for detailed information. ## External Materials See the [Notice and Disclaimer](https://nvidia.github.io/cloudai/#external-materials-notice) in the CloudAI documentation before using external materials retrieved by CloudAI.

جارٍ تحميل ملف README للمستودع…

Skills في هذا المستودع

مطوّرو البرمجيات

١ فئات مهنية · 17% مصنفة

skill

المهنة

الوصف

آخر تحديث

security-assessment

غير مصنف

CloudAI project context and ownership boundaries for security reviews, TAVA, and architecture discussions.

لغة النص الأصلي: الإنجليزية

2026-10-26

cloudai-docs-contribution

غير مصنف

Write, edit, or review CloudAI documentation, including README.md, doc/, release notes, and docstrings.

لغة النص الأصلي: الإنجليزية

2026-09-20

cloudai-remote-experiment

غير مصنف

Run, monitor, stop, and retrieve a CloudAI experiment (test scenario) on a remote Slurm or standalone cluster.

لغة النص الأصلي: الإنجليزية

2026-09-18

cloudai-remote-install

غير مصنف

Install or update a CloudAI checkout on a remote cluster for Slurm or standalone execution. Does not run benchmarks.

لغة النص الأصلي: الإنجليزية

2026-09-18

cloudai-workload-contribution

غير مصنف

Create, modify, or review CloudAI workload implementations. Use for workload Python code, not for TOML-only configuration changes.

لغة النص الأصلي: الإنجليزية

2026-09-18

cloudai-config-contribution

مطوّرو البرمجيات

Create, adapt, review, or troubleshoot CloudAI system, test, and test-scenario TOML configurations. Use for CloudAI config-file requests, not for implementing workload Python code.

لغة النص الأصلي: الإنجليزية

2026-09-16

عرض ٦ من أصل ٦ skills مجمعة.