Sharu Paul

Sharu Paul

Sharu Paul is a Research Scientist at the Genome Informatics Facility at Iowa State University. His background is in Biotechnology with a Ph.D. in Genetics and Genomics. His research is wide-ranging, covering topics from Single-Cell Analyses to the building of Deep Neural Networks. By employing an extensive range of bioinformatic analyses, he seeks to solve big data problems and contributes to developing workshops and tutorials for USDA scientists.

DataScience Workbook / 06: High-Performance Computing (HPC) / 1. Introduction to HPC infrastructure / 1.3. ISU HPC Network / 1.3.2. ISU HPC: Nova computing cluster

Introduction

Nodes and storage can be purchased on the Nova cluster for use by the researchers at ISU.
For details about the Nova cluster, check out the ISU HPC webpage.

Getting Started

Access and Login

Nova login node: @nova.its.iastate.edu Before logging in the first time, check the Loging in section of the Intro to ISU HPC page for information about getting started. Install Google Authenticator on Android ⤴, iPhone ⤴ or Windows ⤴ devices for authentication.

Once connected to ISU network, use any Terminal (Terminal on Linux/Mac and Powershell on Windows) and input following command to log in to Nova:

 ssh <ISU NetID>@nova.its.iastate.edu

Data Transfer and Jobs

Nova transfer node: @novadtn.its.iastate.edu

Use the login hostname @nova.its.iastate.edu to log in to Nova and submit jobs,
but use the data transfer node for data transfers @novadtn.its.iastate.edu

# Login:
ssh Your-ISU-NetID@novadtn.its.iastate.edu

# Copy data:
cp /home/$USER/test.txt /myfiles/Users/$USER/

Once logged in you can start using Nova cluster and submit jobs using SLURM workload manager.

SLURM is used to allocate memory and processors to an input job on Nova.

If not familiar with the SLURM syntax, use this Script generator for Nova ⤴, input your requirements and it will generate a text script. You can then copy this script in a text file, add your commands at the bottom and then submit the job.

Check out the Introduction to SLURM to learn about running your job on Nova or any other HPC Cluster.

#!/bin/bash

#SBATCH --time=01:00:00   # walltime limit (HH:MM:SS)
#SBATCH --nodes=1   # number of nodes
#SBATCH --ntasks-per-node=36   # 36 processor core(s) per node
#SBATCH --mail-user=netid@iastate.edu
#SBATCH --mail-type=begin
#SBATCH --mail-type=end

# LOAD MODULES, INSERT CODE, AND RUN YOUR PROGRAMS HERE
module load modulename
your code/commands

Copy/paste the above job script into a text file, add your modules and commands, and submit with the following command:

sbatch filename

Check your queued jobs:

squeue -u your_netID

Available Software

Check out the following: