Skip to main content

A tool to retrieve and display Slurm usage data

Project description

slurm-scredits

scredits is a Slurm utility for checking account balance. The utility calculates the remaining service units or SU left in the account. The utility shows SU as an aggregate of cpu+gpu+mem usage.

Prerequisites

  • Slurm with Accounting enabled.
  • TRES resources enabled and GrpTRESMins billing set.
  • Optionally gres/gpu enabled and configured.

Usage

usage: scredits [-h] [-v] [-V] [-d] [-a ACCOUNT]

Retrieve and display Slurm usage data.

options:
  -h, --help            show this help message and exit
  -v, --verbose         Print debug messages
  -V, --version         Print program version
  -d, --detailed        Show detailed account and user association
  -a ACCOUNT, --account ACCOUNT
                        Account name to filter results

Installation

pip install scredits

Setting up Slurm

scredits currently support the following setup.

  • Balance is limited per account
  • Account limit is set through GrpTRESMins using billing parameter.

Following is an example setup

Creating account test_account with billing balance of 1000

sacctmgr add account test_account set GrpTRESMins=billing=1000

Add test_user user to account test_account

sacctmgr add user test_user set Account=test_account
sacctmgr add user test_user2 set Account=test_account

Checking balance for all the Accounts

[test@localhost ~]$ scredits
Account         | Allocation(SU)  | Remaining(SU)   | Used(SU)   | Used(%) |
-----------------------------------------------------------------------------
test_account    | 1000.0          | 1000.0          | 0          | 0.0

If you want more details use the -d flag.

[test@localhost ~]$ scredits -d
------------------------------------------------------------------------------------------
Account              | User            | Consumed (SU)   | % SU Usage      | Used Resources
------------------------------------------------------------------------------------------
root                 |                 |                 |                 |
                     | root            | 0               | 0.00%           | cpu=0, mem=0, gpu=0
                     |                 |                 |                 |
                     | Total:          | 0/0             | 0.00%           | cpu=0, mem=0, gpu=0
------------------------------------------------------------------------------------------
test_account         |                 |                 |                 |
                     | test_account    | 0               | 0.00%           | cpu=0, mem=0, gpu=0
                     | test_account2   | 0               | 0.00%           | cpu=0, mem=0, gpu=0
                     |                 |                 |                 |
                     | Total:          | 0/1000          | 0.00%           | cpu=0, mem=0, gpu=0
------------------------------------------------------------------------------------------

You can filter for Account with the -a flag

[test@localhost ~]$ scredits -d -a test_account
------------------------------------------------------------------------------------------
Account              | User            | Consumed (SU)   | % SU Usage      | Used Resources
------------------------------------------------------------------------------------------
test_account         |                 |                 |                 |
                     | test_account    | 0               | 0.00%           | cpu=0, mem=0, gpu=0
                     | test_account2   | 0               | 0.00%           | cpu=0, mem=0, gpu=0
                     |                 |                 |                 |
                     | Total:          | 0/1000          | 0.00%           | cpu=0, mem=0, gpu=0
------------------------------------------------------------------------------------------

Build yourself

Clone the repo

git clone https://github.com/giuliolibrando/slurm-scredits.git

enter into the folder

cd slurm-scredits

install via pip

pip install .

https://pypi.org/project/scredits

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

scredits-1.2.0.tar.gz (6.7 kB view details)

Uploaded Source

Built Distribution

scredits-1.2.0-py3-none-any.whl (7.2 kB view details)

Uploaded Python 3

File details

Details for the file scredits-1.2.0.tar.gz.

File metadata

  • Download URL: scredits-1.2.0.tar.gz
  • Upload date:
  • Size: 6.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.10.12

File hashes

Hashes for scredits-1.2.0.tar.gz
Algorithm Hash digest
SHA256 dd0837fa489a851ed1e09ce48ac5a0a853afd2dd7023b4eee9e30d1f0d0b0dd9
MD5 42804dcf880afbd8cad71238aa67037d
BLAKE2b-256 b183f960ab84b8279f6bfbad1b70df439d5eb99bbe4cf5cf872318f89b5e6056

See more details on using hashes here.

File details

Details for the file scredits-1.2.0-py3-none-any.whl.

File metadata

  • Download URL: scredits-1.2.0-py3-none-any.whl
  • Upload date:
  • Size: 7.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.10.12

File hashes

Hashes for scredits-1.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 da3a2c6bb3fa91348ba631d60d98df9e697018aaae71c9fe114e998aafb5ba3f
MD5 fbbbb7b29d99a6d2c52500ef86344374
BLAKE2b-256 d21bb686164dbf118b4ef1b05c6c8cda0cc77365850d0315da3932a0c988ca6f

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page