I find this a very handy tool not just for multi-user clusters. There are examples for modulefiles out there on the interwebs. You now just need to create some directory containing the so-called modulefiles and add it to $MODULEPATH, for example with module use /path/to/your/modulefiles.
Install pyspark on ubuntu command line how to#
Some users can perhaps comment on how to improve this. I have the feeling that this function is actually not very safe because it evaluates anything you supply it with in its arguments. It effectively runs modulecmd bash plus the arguments you give it. This line defines the module command you were looking for. You need to uncomment the last line and comment out or delete the second-last. It will put a few lines into your shell initalization file, in this case ~/.bashrc: case "$0" in I just did it in under 10 minutes for Ubuntu 14.10 and the Bash shell.Īs mentioned in other posts, you can install Environment Modules from the repos (package environment-modules). Now for something actually Ubuntu-related: It is actually not that much of a pain to set up Environment Modules under Ubuntu. Read more about Environment Modules on HPC here. The instructions you got were probably from some HPC cluster, since you want to run a global climate model. It is a tool for modifying the shell environment (PATH but also aliases etc.) and common to many multi-user clusters like High-Performance Computing clusters. You are looking for the module command provided by the Environment Modules package. Just to summarize the previous comments and answers and add a bit of first-hand installation experience:
0 kommentar(er)
