sreport is a command-line tool used to generate reports on job, user, and cluster-level data from the accounting data collected by the SLURM job scheduler. Accounting data refers to information about the resources used by individual jobs and users on the cluster, such as CPU time, memory usage, and network traffic.
With sreport, administrators can generate reports on various aspects of job and user activity on the cluster, such as the number of jobs submitted by each user, the average CPU time per job, and the most common job states. Additionally, sreport can be used to generate reports on the usage of specific resources, such as CPU cores, memory, and network bandwidth.
sreport offers a wide range of options to filter and sort the report data, including specifying the time period for which to generate the report and filtering based on job or user attributes, such as job state, job ID, user ID, and partition.
If you encounter the below error while running the command sreport:
sreport: command not found
you may try installing the below package as per your choice of distribution:
| Distribution | Command |
|---|---|
| Debian | apt-get install slurm-client |
| Ubuntu | apt-get install slurm-client |
| Kali Linux | apt-get install slurm-client |
| Fedora | dnf install slurm |
| OS X | brew install slurm |
| Raspbian | apt-get install slurm-client |
sreport Command Examples
1. Show pipe delimited cluster utilization data:
# sreport --parsable cluster utilization
2. Show number of jobs run:
# sreport job sizes printjobcount
3. Show users with the highest CPU time use:
# sreport user topuser
Summary
In summary, sreport is a powerful tool for administrators of SLURM-based HPC clusters to generate reports on job and user activity and resource usage. It provides a detailed view of the usage of computing resources on the cluster, which can be used to optimize resource allocation and improve overall cluster performance.