aedavids / bme-237-applied-RNABioinformatics

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Using Hummingbird

Connecting

  • use cisco anyConnect to access ucsc vpn
  • use ssh hb. see ~/.ssh/config

getting started with hummingbird see section about modules

Useful commands

$ module list
$ module avail

# load will modify your PATH. You need to run load every time you login
$ module load hisat/hisat2-2.1.0

export HISAT2_HOME=/hb/software/apps/hisat2/gnu-2.1.0

set up an interactive session on hb using SLURM

weak interactive session. You can not do an big or long running jobs

  1. connect to humming bird
ssh hb
  1. Request a node resource to execute your interactive program
salloc --partition=Instruction --nodes=1 --time=04:00:00 --mem=500M --cpus-per-task=1
salloc: Granted job allocation 94986

# verify
[aedavids@hb ~]$ export | grep SLURM
declare -x SLURM_CLUSTER_NAME="hbhpc"
declare -x SLURM_CPUS_PER_TASK="1"
declare -x SLURM_JOBID="94986"
declare -x SLURM_JOB_CPUS_PER_NODE="1"
declare -x SLURM_JOB_ID="94986"
declare -x SLURM_JOB_NAME="bash"
declare -x SLURM_JOB_NODELIST="hbcomp-005"
declare -x SLURM_JOB_NUM_NODES="1"
declare -x SLURM_JOB_PARTITION="Instruction"
declare -x SLURM_JOB_QOS="normal"
declare -x SLURM_MEM_PER_NODE="500"
declare -x SLURM_NNODES="1"
declare -x SLURM_NODELIST="hbcomp-005"
declare -x SLURM_NODE_ALIASES="(null)"
declare -x SLURM_SUBMIT_DIR="/hb/home/aedavids"
declare -x SLURM_SUBMIT_HOST="hb"
declare -x SLURM_TASKS_PER_NODE="1"
  1. log on to allocated node
 ssh $SLURM_NODELIST
  1. when you are done make sure you use exit to leave slurm interactive shell

About


Languages

Language:Jupyter Notebook 79.4%Language:Shell 20.6%