sprofile

Name	sprofile JSON
Version	0.2.2 JSON
	download
home_page
Summary	Reporting tool for resources actually consumed by a SLURM job.
upload_time	2024-02-12 21:37:48
maintainer
docs_url	None
author
requires_python
license	CECILL-C
keywords	hpc slurm profiling optimization cgroups
VCS
bugtrack_url
requirements	No requirements were recorded.
Travis-CI	No Travis.
coveralls test coverage	No coveralls.

            # SProfile

Sprofile prints the consumed CPU, RAM and GPU resources at the end of a slurm job.
It parses readily available resource usage information and therefore incurs no overhead.

Sprofile can be installed from pypi or from source:

```sh
pip install sprofile
```

For CPU and RAM statistics, slurm must be configured to use the cgroup plugin.
For GPU resource informations, accounting mode must be unabled in the nvidia driver (`nvidia-smi --accounting-mode=1`).

In order to use sprofile, add the following lines at the beginning and the end of the slurm script:

```sh
#!/usr/bin/env sh

...

srun sprofile start

...

srun sprofile stop
```

The last command will print actual resource utilization:

```
-- sprofile report (node27) --
  Time:       0:00:25  /  1:00:00
  CPU load:       0.9  /   2.0
  RAM peak:        3G  /    8G
  GPU load:       0.9  /   1.0
  GPU peak mem:    3G  /   32G
  GPU energy:     0.0kWh
```

            

Raw data

            {
    "_id": null,
    "home_page": "",
    "name": "sprofile",
    "maintainer": "",
    "docs_url": null,
    "requires_python": "",
    "maintainer_email": "",
    "keywords": "hpc,slurm,profiling,optimization,cgroups",
    "author": "",
    "author_email": "Nicolas Granger <nicolas.granger@cea.fr>",
    "download_url": "https://files.pythonhosted.org/packages/e1/df/37a7454b19debfa33fae44045f4dae4a8cc47ae9945316d7059c3598e9c3/sprofile-0.2.2.tar.gz",
    "platform": null,
    "description": "# SProfile\n\nSprofile prints the consumed CPU, RAM and GPU resources at the end of a slurm job.\nIt parses readily available resource usage information and therefore incurs no overhead.\n\nSprofile can be installed from pypi or from source:\n\n```sh\npip install sprofile\n```\n\nFor CPU and RAM statistics, slurm must be configured to use the cgroup plugin.\nFor GPU resource informations, accounting mode must be unabled in the nvidia driver (`nvidia-smi --accounting-mode=1`).\n\nIn order to use sprofile, add the following lines at the beginning and the end of the slurm script:\n\n```sh\n#!/usr/bin/env sh\n\n...\n\nsrun sprofile start\n\n...\n\nsrun sprofile stop\n```\n\nThe last command will print actual resource utilization:\n\n```\n-- sprofile report (node27) --\n  Time:       0:00:25  /  1:00:00\n  CPU load:       0.9  /   2.0\n  RAM peak:        3G  /    8G\n  GPU load:       0.9  /   1.0\n  GPU peak mem:    3G  /   32G\n  GPU energy:     0.0kWh\n```\n",
    "bugtrack_url": null,
    "license": "CECILL-C",
    "summary": "Reporting tool for resources actually consumed by a SLURM job.",
    "version": "0.2.2",
    "project_urls": null,
    "split_keywords": [
        "hpc",
        "slurm",
        "profiling",
        "optimization",
        "cgroups"
    ],
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "blake2b_256": "3ac7372e0098f498d63689f4968ae5631138e9efe47685fae58da57feb8b4823",
                "md5": "ed23440aeacbd7b846f47688b8e2fbf8",
                "sha256": "1e0f9bdbca2c158bcc3d89b305ea3b5bea9ff4a9172e983616ec85e0c7172082"
            },
            "downloads": -1,
            "filename": "sprofile-0.2.2-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "ed23440aeacbd7b846f47688b8e2fbf8",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": null,
            "size": 12343,
            "upload_time": "2024-02-12T21:37:46",
            "upload_time_iso_8601": "2024-02-12T21:37:46.569503Z",
            "url": "https://files.pythonhosted.org/packages/3a/c7/372e0098f498d63689f4968ae5631138e9efe47685fae58da57feb8b4823/sprofile-0.2.2-py3-none-any.whl",
            "yanked": false,
            "yanked_reason": null
        },
        {
            "comment_text": "",
            "digests": {
                "blake2b_256": "e1df37a7454b19debfa33fae44045f4dae4a8cc47ae9945316d7059c3598e9c3",
                "md5": "455c86f64402cf1dbaec51639d9609d2",
                "sha256": "692818d956907de28c645aef6dc46e459bb6ea6e59df33a841ec4fa52f74f711"
            },
            "downloads": -1,
            "filename": "sprofile-0.2.2.tar.gz",
            "has_sig": false,
            "md5_digest": "455c86f64402cf1dbaec51639d9609d2",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 11852,
            "upload_time": "2024-02-12T21:37:48",
            "upload_time_iso_8601": "2024-02-12T21:37:48.725942Z",
            "url": "https://files.pythonhosted.org/packages/e1/df/37a7454b19debfa33fae44045f4dae4a8cc47ae9945316d7059c3598e9c3/sprofile-0.2.2.tar.gz",
            "yanked": false,
            "yanked_reason": null
        }
    ],
    "upload_time": "2024-02-12 21:37:48",
    "github": false,
    "gitlab": false,
    "bitbucket": false,
    "codeberg": false,
    "lcname": "sprofile"
}