{"id":13521790,"url":"https://github.com/nebuly-ai/nos","last_synced_at":"2025-05-16T18:03:53.861Z","repository":{"id":37597276,"uuid":"480818682","full_name":"nebuly-ai/nos","owner":"nebuly-ai","description":"Module to Automatically maximize the utilization of GPU resources in a Kubernetes cluster through real-time dynamic partitioning and elastic quotas - Effortless optimization at its finest!","archived":false,"fork":false,"pushed_at":"2024-04-21T20:57:30.000Z","size":10252,"stargazers_count":655,"open_issues_count":26,"forks_count":36,"subscribers_count":13,"default_branch":"main","last_synced_at":"2025-04-12T16:52:32.280Z","etag":null,"topics":["gpu","kubernetes","optimization"],"latest_commit_sha":null,"homepage":"https://www.nebuly.com/","language":"Go","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/nebuly-ai.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2022-04-12T13:18:09.000Z","updated_at":"2025-04-11T18:29:30.000Z","dependencies_parsed_at":"2024-01-09T00:56:40.158Z","dependency_job_id":"fde058fa-e63a-4018-9ee9-cf2943e331be","html_url":"https://github.com/nebuly-ai/nos","commit_stats":null,"previous_names":["nebuly-ai/nebulgym"],"tags_count":3,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nebuly-ai%2Fnos","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nebuly-ai%2Fnos/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nebuly-ai%2Fnos/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nebuly-ai%2Fnos/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/nebuly-ai","download_url":"https://codeload.github.com/nebuly-ai/nos/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":254582902,"owners_count":22095518,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["gpu","kubernetes","optimization"],"created_at":"2024-08-01T06:00:38.047Z","updated_at":"2025-05-16T18:03:53.809Z","avatar_url":"https://github.com/nebuly-ai.png","language":"Go","funding_links":[],"categories":["Open Source Projects","Go","Tools and Libraries","Training Orchestration","Optimization Tools"],"sub_categories":["Data Processing and Machine Learning"],"readme":"# Nebuly Operating System (nos)\n\n![](docs/en/docs/img/nos-logo.png)\n\n---\n\n**Documentation**: \u003ca href=\"https://nebuly-ai.github.io/nos/overview\" target=\"_blank\"\u003e docs.nebuly.com/nos/overview \u003c/a\u003e\n\nIf you like the project please support it by leaving a star ✨\n\n---\n\n`nos` is the open-source module to efficiently run AI workloads on Kubernetes,\nincreasing GPU utilization, cutting down infrastructure costs and improving workloads performance.\n\nCurrently, the available features are:\n\n* [Dynamic GPU partitioning](https://nebuly-ai.github.io/nos/dynamic-gpu-partitioning/overview): allow to schedule Pods requesting\nfractions of GPU. GPU partitioning is performed automatically in real-time based on the Pods pending and running in\nthe cluster, so that Pods can request only the resources that are strictly necessary and GPUs are always fully utilized.\n\n* [Elastic Resource Quota management](https://nebuly-ai.github.io/nos/elastic-resource-quota/overview): increase the number of Pods running on the\ncluster by allowing namespaces to borrow quotas of reserved resources from other namespaces as long as they are\nnot using them.\n\n![](docs/en/docs/img/gpu-utilization.png)\n\n\n## Getting started\n\n### Prerequisites\n\n* Kubernetes v1.23 or newer\n* [GPU Support must be enabled](http://nebuly-ai.github.io/nos/prerequisites/#enable-gpu-support)\n* [Nebuly k8s-device-plugin](https://github.com/nebuly-ai/k8s-device-plugin) (optional, required only if you want to enable MPS partitioning)\n* [cert-manager](https://cert-manager.io/docs/) (optional, but recommended)\n\n\n### Installation\n\nYou can install `nos` using Helm 3 (recommended).\nYou can find all the available configuration values in the Chart [documentation](https://nebuly-ai.github.io/nos/helm-charts/nos/).\n\n```bash\nhelm install oci://ghcr.io/nebuly-ai/helm-charts/nos \\\n  --version 0.1.2 \\\n  --namespace nebuly-nos \\\n  --generate-name \\\n  --create-namespace\n```\n\nAlternatively, you can use Kustomize by cloning the repository and running `make deploy`.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnebuly-ai%2Fnos","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnebuly-ai%2Fnos","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnebuly-ai%2Fnos/lists"}