Difference between revisions of "SW:Modules"
m (→HPRC Shorthand)
|Line 77:||Line 77:|
Use -f to force an update.
Revision as of 12:07, 5 January 2021
The Modules System
The Modules system organizes the multitude of packages we have installed on our clusters so that they can be easily maintained and used. More specifically, Modules configure appropriately the execution environment of each package (and version).
Most non-OS software is organized by the Modules system. This includes all compilers and associated libraries. Each package has a corresponding modulefile or simply a module where the appropriate
actions are specified (in a prescibed format) in setting up its execution environment.
Notify the help desk if you cannot locate the module appropriate for an application.
To find out what software packages are available under the Modules system use one of the following two commands:
[NetID@cluster ~]$ module spider [packageName]
[NetID@cluster ~]$ module avail [packageName]
In order to use some software on our clusters, the module for the software must be loaded first. To load the module for a software package, use the following command:
[NetID@cluster ~]$ module load packageName/version
Note: The specification of packageName in the load command is case sensitive and it must include a specific version. No packages are preloaded by default.
Note: Loading a package is required in order to access the man pages associated with the package if they are available.
To find out what packages/modules you have loaded on your current session, use the following command:
[NetID@cluster ~]$ module list
To remove one module from your current session, use the following command:
[NetID@cluster ~]$ module unload [packageName]
To remove ALL modules from your current session, use the following command:
[NetID@cluster ~]$ module purge
Note: It is very important to remove all modules when switching between software and compilers. Mixing modules and versions is NOT a good idea and can cause many problems.
To search and list what packages/modules are available on Ada, enter any suitable combination of the following:
[ NetID@ada ~]$ module spider [packageName] # List all versions of all packages or just for packageName. [ NetID@ada ~]$ module spider string # List all modules that contain the "string". [ NetID@ada ~]$ module spider packageName/version # List detailed information about that version of packageName [ NetID@ada ~]$ module keyword string # List all modules and whatis that contain string.
For additional help see:
[ NetID@ada ~]$ module help # Lists all the subcommands under the module command. [ NetID@ada ~]$ module help packageName # Lists information about packageName
To remove (unload) a module and load another, use the swap subcommand:
[ NetID@ada ~]$ module swap packageOUT packageIN
The above command is short for:
[ NetID@ada ~]$ module unload packageOUT [ NetID@ada ~]$ module load packageIN
A handy front end for the HPRC module system is provided known as ml.
Note: The abbreviation ml can be used instead of module, module load, or module list depending on the situation. This can be seen in the examples below.
[ NetID@ada ~]$ ml means: module list [ NetID@ada ~]$ ml foo bar means: module load foo bar [ NetID@ada ~]$ ml -foo -bar baz goo means: module unload foo bar; module load baz goo;
Any module subcommand name can be given after ml:
if name is a subcommand like avail, save, restore, show, swap,... [ NetID@ada ~]$ ml name arg1 arg2 ... All options must go before the subcommand: [ NetID@ada ~]$ ml --terse avail Then this is the same : [ NetID@ada ~]$ module name arg1 arg2 ...
In other words you can not load a module named: show swap etc.
Use -f to force an update.
[ NetID@ada ~]$ ml -f
The output below shows the output of the spider command for ABAQUS:
[ NetID@ada ~]$ ml spider ABAQUS # module abbreviated as ml ---------------------------------------------------------------------------------------- ABAQUS: ---------------------------------------------------------------------------------------- Description: Finite Element Analysis software for modeling, visualization and best-in-class implicit and explicit dynamics FEA. - Homepage: http://www.simulia.com/products/abaqus_fea.html Versions: ABAQUS/6.12.1-linux-x86_64 ABAQUS/6.13.5-linux-x86_64 . . .
The output below illustrates the fact that a modulefile's complete name includes its version. An installed application can have several versions.
[ NetID@cluster ~]$ ml ABAQUS/2017 # module load abbreviated as ml [ NetID@cluster ~]$ ml # module list abbreviated as ml Currently Loaded Modules: 1) ABAQUS/2017
The output below shows the functionality of the swap command:
[ NetID@cluster ~]$ ml load ABAQUS/2016 [ NetID@cluster ~]$ ml swap ABAQUS/2016 ABAQUS/2017 The following have been reloaded with a version change: 1) ABAQUS/2016 => ABAQUS/2017 [ NetID@cluster ~]$ module list Currently Loaded Modules: 1) ABAQUS/2017
WARNING: Do not include module commands in your startup files!!!
We advise users to NOT put module load commands in their startup files like $HOME/.bashrc and $HOME/.bash_profile. Although you may think this makes life easier you will find in time that it causes all kinds of problems later on (e.g. when switching to different programs or running batch jobs).
If you really want to be able to load a set of modules quickly then look at Lmod's "user collections". You can use module save to create a default collection of modules and then you simply have to run module restore to restore them (BUT DON'T DO THAT in .bashrc/.bash_profile). You can also named collections... e.g. I might do module save openfoam after loading all the OpenFOAM modules I need (after a module purge) and then do a module save ANSYS that has the modules I use with ANSYS. Then I can restore either of those collections with module restore <name>.