This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
|
survival_guide [2020/05/26 12:54] agill [Software Modules on Lengau] |
survival_guide [2022/02/01 18:38] (current) alopis |
||
|---|---|---|---|
| Line 8: | Line 8: | ||
| - The computers that make up the HPC cluster do not run faster than a good desktop computer, nor do they usually have particularly impressive amounts of memory. | - The computers that make up the HPC cluster do not run faster than a good desktop computer, nor do they usually have particularly impressive amounts of memory. | ||
| - The system is designed to be used in batch mode. The user prepares a job, submits it to the queue, and the system executes it when the requested amount of resources become available. | - The system is designed to be used in batch mode. The user prepares a job, submits it to the queue, and the system executes it when the requested amount of resources become available. | ||
| - | - You can request an interactive session, which will give you access to a compute node in the cluster. | + | - You can request an interactive session, which will give you access to a compute node in the cluster. |
| - Access to the system is by means of secure-shell (ssh). | - Access to the system is by means of secure-shell (ssh). | ||
| - Data can be copied to or from the system by means of secure copy (scp) or rsync. | - Data can be copied to or from the system by means of secure copy (scp) or rsync. | ||
| Line 19: | Line 19: | ||
| ===== The user interface (not) ===== | ===== The user interface (not) ===== | ||
| - | Although personal computers running Linux have user interfaces that are very similar to Windows or Mac, the ssh-login to the HPC cluster runs in command line text mode only, by default using the Bash shell. | + | Although personal computers running Linux have user interfaces that are very similar to Windows or Mac, the ssh-login to the HPC cluster runs in command line text mode only, by default using the Bash shell. |
| - | * ls - provide a list of the files in the current directory | + | |
| - | * ls -l - provide a more detailed listing of the files in the current directory | + | |
| - | * pwd - gives the current directory path (Present Working Directory) | + | |
| - | * mkdir - create a new directory | + | |
| - | * cd - change to a new directory | + | |
| - | * rmdir - remove a directory | + | |
| - | * rm - remove a file | + | * **mv** - move a file (either rename it or cut/paste it into a directory). |
| - | * nano - start the nano text editor (recommended for beginners) | + | * **rmdir** - remove a directory |
| - | * vi - start the vi text editor (warning: steep learning-curve) | + | |
| + | | ||
| + | | ||
| + | * **qsub** - submit a batch job to PBS for processing | ||
| + | * **qstat -u $USER** - list the status of your PBS jobs | ||
| + | * **qdel // | ||
| ==== Software Modules on Lengau ==== | ==== Software Modules on Lengau ==== | ||
| Line 34: | Line 39: | ||
| Here are examples of the most common module operations that you will need to use on the cluster: | Here are examples of the most common module operations that you will need to use on the cluster: | ||
| * **module avail** - To list all available packages | * **module avail** - To list all available packages | ||
| + | * **module avail 2>&1 | grep // | ||
| * **module avail chpc** - List just the chpc modules group - this contains most of the useful modules | * **module avail chpc** - List just the chpc modules group - this contains most of the useful modules | ||
| * **module avail chpc/ | * **module avail chpc/ | ||
| Line 39: | Line 45: | ||
| * **module load chpc/ | * **module load chpc/ | ||
| * **module list** - To list all the modules which you have already loaded in this session | * **module list** - To list all the modules which you have already loaded in this session | ||
| - | * **module | + | * **module |
| * **module purge** - to unload all modules that you have loaded | * **module purge** - to unload all modules that you have loaded | ||
| Line 46: | Line 52: | ||
| There are a great number of modules for use by researchers in the life sciences. Some are also generally useful, such as compilers, etc. These are accessible as follows: | There are a great number of modules for use by researchers in the life sciences. Some are also generally useful, such as compilers, etc. These are accessible as follows: | ||
| module avail chpc/ | module avail chpc/ | ||
| + | Modules for most other disciplines/ | ||
| + | * Computational mechanics modules (CFD, FEA, etc) are all in chpc/ | ||
| + | * Earth-science modules are in chpc/earth | ||
| + | * Some physics modules are in chpc/phys | ||
| + | * Most of the computational chemistry and material science software is available directly under chpc (check chpc/ | ||
| For python users, loading a module will also load all the python libraries associated with that python version - usually at least numpy and scipy. The most capable python modules are those in chpc/ | For python users, loading a module will also load all the python libraries associated with that python version - usually at least numpy and scipy. The most capable python modules are those in chpc/ | ||
| Line 51: | Line 62: | ||
| For GCC and the Intel compilers, however, you will have to load the modules that you wish to link your code against separately. | For GCC and the Intel compilers, however, you will have to load the modules that you wish to link your code against separately. | ||
| - | Finally, there are a few software packages (mostly those that are commercially licensed, such as Ansys/ | + | Finally, there are a few software packages (mostly those that are commercially licensed, such as Ansys/ |
| + | ==== Installing New Software on Lengau ==== | ||
| + | First please check that the software which you wish to use is not available via the module system (see above) or some other means. | ||
| + | If it is not available there, your two options are to either ask a CHPC staff member to help you install it, or to try to compile it yourself (not recommended for beginners). | ||
| + | YOU WILL NOT BE ABLE TO INSTALL SOFTWARE AS ON LINUX ON YOUR LAPTOP - EG. "sudo apt-get install something-0.0.1" | ||
| + | * As a user, you don't have sudo permission, and will not be granted it under any circumstances - as a national HPC facility, this would represent an unacceptable security risk. __This is not negotiable.__ | ||
| + | * apt-get install or apt-install are specific to debian-based linux systems, which are popular for laptops, but less so for HPC. Lengau runs a different version of linux called CentOS. | ||
| + | * the software available via the debian or ubuntu repositories is not optimised for HPC systems, which have special hardware configurations, | ||
| + | |||
| + | The CHPC staff will do their best to assist you by installing the software that you request. However this may take considerable time and effort for certain pieces of software to be installed, so you may need to be patient. HPC systems are complex, so unfortunately the installation process is often also complex and tricky. Also, you should ensure that the software which you wish to use will actually benefit from being used on an HPC system. As mentioned already, some software is not written to take advantage of HPC. | ||
| + | |||
| + | If you wish to use commercially licensed software, you will need to have an appropriate license. Usually this will require a 3-way communication between you, the software licensor and the CHPC. The CHPC does have a small number of national licenses for very popular pieces of software, but is unlikely to procure new licenses of this sort at this time. | ||