Mana (at University of Hawaii at Manoa) Configuration
To use, run the pipeline with -profile mana. It will use the following parameters for Mana (UHM HPCC):
- Load singularity and use it as default container technology
- Setup a container cache directory in your home (~/.singularity_images_cache)
- Select appropriate queues (currently:
shared,exclusive,kill-shared,kill-exclusive) - Set the maximum available resources (available in 09/02/2022):
- CPUs: 96
- Memory: 400.GB
- Time: 72.h
Pre-requisites
In order to run a nf-core pipeline on Mana, you will need to setup nextflow in your environment. At the moment, nextflow is not available as a module (but might be in the future).
Install nextflow in a conda environment
Before we start, we will need to work on an interactive node (currently, mana doesn’t let you execute any program in the login node):
# Request an interactive sandbox node for 30 min
srun --pty -t 30 -p sandbox /bin/bashTo setup nextflow on your account, follow these steps.
# Load the latest anaconda3 module
module load lang/Anaconda3/2022.05
# Initialize environment
. $(conda info --base)/etc/profile.d/conda.sh
# Install nextflow (here in base environment, but you can create a new one if you'd like)
conda install -c bioconda nextflowIf you want these settings to be persistent, you can add the first 2 commands in your .bash_profile file like this:
echo "module load lang/Anaconda3/2022.05" >> ~/.bash_profile
echo "$(conda info --base)/etc/profile.d/conda.sh" >> ~/.bash_profileConfig file
params {
config_profile_description = 'University of Hawaii at Manoa'
config_profile_url = 'http://www.hawaii.edu/its/ci/'
config_profile_contact = 'Cedric Arisdakessian'
max_memory = 400.GB
max_cpus = 96
max_time = 72.h
}
process {
resourceLimits = [
memory: 400.GB,
cpus: 96,
time: 72.h
]
executor = 'slurm'
queue = 'shared,exclusive,kill-shared,kill-exclusive'
module = 'tools/Singularity'
}
singularity {
enabled = true
cacheDir = "${HOME}/.singularity_images_cache"
autoMounts = true
}
executor
slurm
contact
