From ALICE Documentation
Frequently Asked Questions
Questions about logging in
Your account name is the same as your University Account name, but it is a separate account with its own password. Provide this ULCN account name if you request an ALICE account.
Your account will not expire until your ULCN account does. In case of any doubt please contact email@example.com
You can follow the instructions on SSH Keys
There are multiple ways to access Alice. Choose your preferred method here
You will need access to the University SSH gateway. See SSH gateway
When you login at the Alice login node, by default the file .bashrc will be run. You can add any command that you want to run at login to this file.
Questions about using Alice
See: Load modules
Contact firstname.lastname@example.org. Since software you need may also be useful to other ALICE users we will install most packages in the module repository for you. Please expect a little delay time after being honoured the central install as sometimes the install process can be lengthy and we have to make sure it is not clashing with other already installed packages.
Questions about Files and Sharing
You have been entered in a specific UNIX group. This group has the name of your institute and you can share files with all your ALICE institute colleagues by running the Unix change mod. utility: chmod g+r on files and chmod g+x on directories.
Make sure that you use the data share, which is available at /data, instead of your home directory. See also: Data Storage
Your data can be stored a limited amount of time. ALICE is not meant to be a storage facility and thus you will require other facilities to store and backup your data. You will be contacted if you have old data that is in the way of new users requirements. Disk space is limited so be a gentle user.
Questions about Slurm
You can find all information about submitting jobs here
Canceling a job is straight forward. With
squeue you can lookup the ID of the job you want to cancel and the run the
scancel command with the job ID. See here for details.
You are not the only person running jobs on the cluster. It might take some time for your job to be granted the required resources before it will actually start executing. For more information about Job Resource see the wile page at Job Resources. You may also accidentily be using high data volumes from the
/data central storage facility instead of local node storage. For more information about storge see the wiki page about Data Storage