Différences
Ci-dessous, les différences entre deux révisions de la page.
| Les deux révisions précédentes Révision précédente Prochaine révision | Révision précédente | ||
| quick_start [2018/02/06 13:24] – admin | quick_start [2018/06/18 14:45] (Version actuelle) – admin | ||
|---|---|---|---|
| Ligne 2: | Ligne 2: | ||
| This document is intended for new users of the service. It gives information on such basics as how to log in, how to run application and where to get help. | This document is intended for new users of the service. It gives information on such basics as how to log in, how to run application and where to get help. | ||
| + | |||
| + | **Computation Workflow on the HPC** | ||
| + | |||
| + | A typical Computation workflow will be: | ||
| + | - Connect to the HPC | ||
| + | - Transfer your files to the HPC | ||
| + | - Compile your code and test it | ||
| + | - Create a job script | ||
| + | - Submit your job | ||
| + | - Wait while | ||
| + | * your job gets into the queue | ||
| + | * your job gets executed | ||
| + | * your job finishes | ||
| + | - Move your results | ||
| + | |||
| ====== Connecting to clusters ====== | ====== Connecting to clusters ====== | ||
| Ligne 15: | Ligne 30: | ||
| </ | </ | ||
| - | Example | + | Example, connect and change password: |
| <code bash> | <code bash> | ||
| Ligne 40: | Ligne 55: | ||
| Once connected each user will have a home directory organized as follow : | Once connected each user will have a home directory organized as follow : | ||
| - | * '' | + | * '' |
| - | * '' | + | * '' |
| Read more [[environnement_travail|here]] | Read more [[environnement_travail|here]] | ||
| Ligne 48: | Ligne 63: | ||
| < | < | ||
| - | Several software are installed on the clusters | + | Several software are installed on the clusters. You can show them by : |
| <code bash> | <code bash> | ||
| $ module avail | $ module avail | ||
| Ligne 54: | Ligne 69: | ||
| This will show the list of software installed and their version. | This will show the list of software installed and their version. | ||
| - | the output, on mesocluster, | + | the output looks like this : |
| <code bash> | <code bash> | ||
| Ligne 152: | Ligne 167: | ||
| Where '' | Where '' | ||
| - | These files will be in the directory you submitted the jobs from if you use the -cwd option, and in your home directory otherwise. The names are constructed from the name of the job script, '' | + | These files will be stored |
| <note important> | <note important> | ||
| JOB must bhe launched from WORK directory. | JOB must bhe launched from WORK directory. | ||
| Ligne 231: | Ligne 246: | ||
| ==== MPI script ==== | ==== MPI script ==== | ||
| <note tip> | <note tip> | ||
| - | To use MPI, you need to specify that you wish to use that Parallel Environment (-pe impi_tight), and declare the number of slots</ | + | To use MPI, you need to specify that you wish to use that Parallel Environment (-pe mpi), and declare the number of slots</ |