Difference between revisions of "SciNet Command Line Utilities"
From SciNet Users Documentation
(→Jobs & Queues) |
(→Storage) |
||
Line 21: | Line 21: | ||
|''-h'' to see all command options | |''-h'' to see all command options | ||
|Informs about the user and group file system usage. | |Informs about the user and group file system usage. | ||
− | |Niagara | + | |Niagara, BGQ |
|- | |- | ||
|topUserDirOver1000list | |topUserDirOver1000list |
Revision as of 01:07, 10 August 2018
Below is list of handy command-line utilities written by SciNet to help manage your data and compute.
Note that in the table, optional arguments are denoted with square brackets (i.e. '[]'), while mandatory arguments are denoted with angular brackets (i.e. '<>').
Storage
Command | Arguments | Description | Cluster |
---|---|---|---|
quota | Short overview of a user's storage usage. | Niagara | |
diskUsage | -h to see all command options | Informs about the user and group file system usage. | Niagara, BGQ |
topUserDirOver1000list | Lists your directories that have over 1,000 files | Niagara | |
topUserDirOver1GBlist | Lists your directories that have over 1 GB of data | Niagara |
Jobs & Queues
Command | Arguments | Description | Cluster |
---|---|---|---|
qsum | most squeue arguments work | Lists jobs running or in the queue, grouped by user. | Niagara |
llq1 | Detailed information on jobs that are actively running. | TCS | |
scinet gpc priority | Compute usage in the last 14 days, and how this affects your priority in the queue. | GPC | |
nodeperf | [userName] | Who is doing what on the current node? | GPC |
jobperf | <jobId | jobName> | Informs about the performance of all nodes of a given job. | GPC |
jobError | <jobId | jobName> | Displays realtime error output of a given job. | GPC |
jobOutput | <jobId | jobName> | Displays realtime standard output of a given job. | GPC |
jobcd | <jobId | jobName> | Allows users to quickly move into the working directory of a given job. | GPC |
jobscript | <jobId | jobName> | Displays the submission script used when submitting a given job. | GPC |
jobssh | <jobId | jobName> | Allows users to connect to the head-node of a given job. | GPC |
jobtop | <jobId | jobName> | Allows users to "top" on the head-node of a given job. | GPC |
jobdep | <jobId> | Displays dependencies of given job. | GPC |
jobtree | [userName] | Displays the jobs tree of dependencies for a given user. | GPC |
debugjob | [nodes] | Requests a time-limited interactive session of up to 8 dedicated nodes | GPC |
Modules
Template:Hl2 | Command | Template:Hl2| Arguments | Template:Hl2| Description | Template:Hl2| Cluster |
---|---|---|---|
modulefind | [searchWord] | Finds modules containing searchWord and indicate their compatibility with already loaded modules. | GPC |
moduleadvice | <modulename> | Show the module commands required to load modulename, taking into account prerequisites and conflicts with already loaded modules. | GPC |
modulewrap | <comma-separated-modules> <command> | Load the modules in the comma-separated-modules list (no spaces!) only to run the given command. | GPC |
Related Topics
Checking the memory usage from jobs
GPC Quickstart
Scheduler
Monitoring Jobs
Tech Talk on Monitoring Jobs