{"id":17344145,"url":"https://github.com/typhoonzero/fluid_gpu_benchmark","last_synced_at":"2025-10-13T01:20:24.005Z","repository":{"id":99429027,"uuid":"125504814","full_name":"typhoonzero/fluid_gpu_benchmark","owner":"typhoonzero","description":null,"archived":false,"fork":false,"pushed_at":"2018-06-25T06:32:22.000Z","size":14944,"stargazers_count":1,"open_issues_count":0,"forks_count":1,"subscribers_count":0,"default_branch":"master","last_synced_at":"2025-10-04T18:45:24.816Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/typhoonzero.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2018-03-16T11:02:17.000Z","updated_at":"2018-06-25T06:32:23.000Z","dependencies_parsed_at":"2023-09-09T12:01:20.282Z","dependency_job_id":null,"html_url":"https://github.com/typhoonzero/fluid_gpu_benchmark","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/typhoonzero/fluid_gpu_benchmark","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/typhoonzero%2Ffluid_gpu_benchmark","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/typhoonzero%2Ffluid_gpu_benchmark/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/typhoonzero%2Ffluid_gpu_benchmark/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/typhoonzero%2Ffluid_gpu_benchmark/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/typhoonzero","download_url":"https://codeload.github.com/typhoonzero/fluid_gpu_benchmark/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/typhoonzero%2Ffluid_gpu_benchmark/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":279013886,"owners_count":26085325,"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","status":"online","status_checked_at":"2025-10-12T02:00:06.719Z","response_time":53,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":[],"created_at":"2024-10-15T16:24:13.564Z","updated_at":"2025-10-13T01:20:23.990Z","avatar_url":"https://github.com/typhoonzero.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Fluid Benchmark\n\nThis directory contains several models and tools that used to run\nFluid benchmarks for local and distributed training.\n\n\n## Run the Benchmark\n\nTo start, run the following command to get the full help message:\n\n```bash\npython fluid_benchmark.py --help\n```\n\nCurrently supported `--model` argument include:\n\n* mnist\n* resnet\n    * you can chose to use different dataset using `--data_set cifar10` or\n      `--data_set flowers`.\n* vgg\n* stacked_dynamic_lstm\n* machine_translation\n\n* Run the following command to start a benchmark job locally:\n    ```bash\n      python fluid_benchmark.py --model mnist --parallel 1 --device GPU --with_test\n    ```\n    You can choose to use GPU/CPU training. With GPU training, you can specify\n    `--parallel 1` to run multi GPU training.\n* Run distributed training with parameter servers:\n    * start parameter servers:\n        ```bash\n        PADDLE_TRAINING_ROLE=PSERVER PADDLE_PSERVER_PORT=7164 PADDLE_PSERVER_IPS=127.0.0.1 PADDLE_TRAINERS=1 PADDLE_CURRENT_IP=127.0.0.1 PADDLE_TRAINER_ID=0 python fluid_benchmark.py --model mnist --parallel 0 --device GPU --update_method pserver\n        ```\n    * start trainers:\n        ```bash\n        PADDLE_TRAINING_ROLE=PSERVER PADDLE_PSERVER_PORT=7164 PADDLE_PSERVER_IPS=127.0.0.1 PADDLE_TRAINERS=1 PADDLE_CURRENT_IP=127.0.0.1 PADDLE_TRAINER_ID=0 python fluid_benchmark.py --model mnist --parallel 0 --device GPU --update_method pserver\n        ```\n* Run distributed training using NCCL2\n    ```bash\n    PADDLE_PSERVER_PORT=7164 PADDLE_TRAINER_IPS=192.168.0.2,192.168.0.3  PADDLE_CURRENT_IP=127.0.0.1 PADDLE_TRAINER_ID=0 python fluid_benchmark.py --model mnist --parallel 0 --device GPU --update_method nccl2\n    ```\n\n## Run Distributed Benchmark on Kubernetes Cluster\n\nWe provide a script `kube_gen_job.py` to generate Kubernetes yaml files to submit\ndistributed benchmark jobs to your cluster. To generate a job yaml, just run:\n\n```bash\npython kube_gen_job.py --jobname myjob --pscpu 4 --cpu 8 --gpu 8 --psmemory 20 --memory 40 --pservers 4 --trainers 4 --entry \"python fluid_benchmark.py --model mnist --parallel 1 --device GPU --update_method pserver --with_test\" --disttype pserver\n```\n\nThen the yaml files are generated under directory `myjob`, you can run:\n\n```bash\nkubectl create -f myjob/\n```\n\nThe job shall start.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftyphoonzero%2Ffluid_gpu_benchmark","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftyphoonzero%2Ffluid_gpu_benchmark","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftyphoonzero%2Ffluid_gpu_benchmark/lists"}