1. Beginner guide

This is a step by step beginner guide that explains how to get an account, to login, and to do a few simple things on HPC3 cluster.

1.1. Your Laptop

You will need to have a few applications on your laptop, most are standard:

Table 1.5 Your laptop applications

Application

Type

Your laptop model

Linux

macOS

Windows

Windows 10

VPN access

Install client VPN software per the instructions UCI campus VPN

Terminal

use your

favorite

Terminal or

iTerm2

PuTTY or

MobaXterm

Windows Terminal, MobaXterm

or Linux Subsystem for Windows

Secure Connect

ssh

ssh

ssh

Secure Data

Transfer

scp

or rsync

scp

or rsync

Filezilla or

WinSCP

scp or rsync

(in Linux Subsystem for Windows)

Attention

MobaXterm users DO NOT enable Remote monitoring! See MobaXterm for more info.

1.2. Login

Before you login please make sure you have an account. Please see Getting an account.
Once you have an account you can login to HPC3 cluster.
HPC3 is a shared facility. Please read Acceptable use guide.

The HPC3 cluster name is hpc3.rcic.uci.edu and it should be used in all connections.

We require multi-factor authentication for all password-based logins. Please make sure you have enabled your DUO device using UCI’s Duo infrastructure.

To directly login on HPC3 cluster you must:

  • be connected to the UCI campus VPN (or be on the campus network)

  • use ssh command from your Terminal application

  • have a way to use DUO authentication

Step 1

Connect to UCI campus VPN, see UCI campus VPN instructions.

Step 2

Open your Terminal application on your laptop and start ssh session to connect to the cluster login node. Use a cluster name and your your regular UCI credentials (UCInetID and password). For example, a user with UCInetID panteater will use:

$ ssh panteater@hpc3.rcic.uci.edu

When prompted for a password enter your UCInetID password followed by Return key. Note, password will not be visible when typed:

Password:
Step 3

After that, user will get a DUO prompt to enter a passcode or request a push to the enrolled DUO-enabled device (user’s phone). A prompt looks similar to:

Duo two-factor login for panteater

Enter a passcode or select one of the following options:

 1. Duo Push to XXX-XXX-1234

Passcode or option (1-1):

Type desired option (in this example 1):

Passcode or option (1-1): 1

Now use the DUO app on your phone and respond to the received DUO notification by pressing Approve when prompted. If the DUO authentication is successful you will see in your laptop window:

Success. Logging you in...

And after a successful login you will see a screen similar to the following:

+-----------------------------------------+
|  _             _             _ _ ____   |
| | | ___   __ _(_)_ __       (_) | ___|  |
| | |/ _ \ / _` | | '_ \ _____| | |___ \  |
| | | (_) | (_| | | | | |_____| | |___) | |
| |_|\___/ \__, |_|_| |_|     |_|_|____/  |
|          |___/                          |
+-----------------------------------------+
 Distro:  Rocky 8.7 Green Obsidian
 Virtual: NO

 CPUs:    40
 RAM:     191.8GB
 BUILT:   2022-08-30 14:02

 ACCEPTABLE USE: https://rcic.uci.edu/documents/RCIC-Acceptable-Use-Policy.pdf

 [panteater@login-i15:~]$

The above text output screen is called motd. It includes a general information about the cluster login node (we have a few) plus important messages about the cluster such as pending shutdowns or links to important online documents.

The last line of the output is your shell prompt (yours may look slightly different), this is where you can type commands.

1.3. File editing

Users will need to learn one of file editors: vim or emacs. Choose the editor that is more intuitive for you. See the File Editors beginner tutorials, many more are available online.

Important

Please avoid using Special Characters in file and directory names.

1.4. Simple commands

Users who are unfamiliar with Linux environment will need to learn the basics of bash shell, file editing, or using language such as R or Python. Please see the Tutorials page that lists links to various beginner guides.

The cluster shell bash is a command language interpreter that executes commands that you type (the standard input). Bash shell automatically provides a prompt which is a text similar to [user@login-x:~]$ and is usually the username followed by the hostname. You don’t need to type prompt.

Below is a small set of simple but very useful commands to try. What you type is immediately after the prompt [user@login-x:~]$. Each command returns an output that will be displayed in your terminal window and will be similar to the following:

Print name of current/working directory:

[user@login-x:~]$ pwd
/data/homezvol0/panteater

Print current date and time in default format:

[user@login-x:~]$ date
Mon May 19 12:43:42 PDT 2023

Print host name:

The cluster has a few login nodes and multiple working nodes, each has its own unique name.

[user@login-x:~]$ hostname
login-i15

List current directory contents:

[user@login-x:~]$ ls
local

Long list using -l option:

Lists all directory contents including hidden files that start with dot . and info about each file.

[user@login-x:~]$ ls -l
drwx------   3 panteater panteater    9 Jul 13 00:02 .
drwxr-xr-x 785 root      root       785 Jul 16 10:32 ..
-rw-r--r--   1 panteater panteater  183 Jul 12 14:42 .bash_profile
-rw-r--r--   1 panteater panteater  541 Jul 12 14:42 .bashrc
-rw-r--r--   1 panteater panteater  500 Jul 12 14:42 .emacs
-rw-r--r--   1 panteater panteater   17 Jul 12 14:42 .forward
drwxrwxr-x   2 panteater panteater    2 Jun 15 09:48 local

By default, many commands need no arguments or additional flags. Arguments given to the commands provide more specific information in the output, as the last command ls -l did.

To learn about specific commands consult tutorials or use manual pages via man command. For example to learn more about ls command type:

[user@login-x:~]$ man ls

and use the space key to scroll through the output on the screen.

1.5. Running applications

Please follow Acceptable use to avoid problems. Cluster is a shared resource, at any time there are many users logged in and hundreds of jobs are running. What you do can adversely affect others.

The cluster environment is not well suited for GUI type of applications. Most of the commands users need to type in, there are no clickable icons and no pop-up windows.

We use Slurm scheduler to run CPU-intensive or long running applications. Slurm guide provides extensive info about using the scheduler. Here is a brief summary:

  • Slurm is an open-source workload manager for Linux clusters that manages and schedules jobs on the cluster.

  • HPC3 has different kinds of nodes (servers) that are separated into groups according to their resources (memory, CPU, etc). Slurm uses the term partition to signify a group of such resources. All jobs are submitted to the nodes in partitions.

We have a few partitions, most users will need to use:

  • free partition is for jobs that can be preempted (killed) by standard jobs. There are no charges for using this partition.

  • standard partition is for jobs that should not be interrupted. Usage is charged against:

    • The user’s Slurm bank account. Each user gets FREE one time allocation of 1000 core hours to run jobs here. Users are NOT CHARGED FOR this allocation, but it can not be extended or renewed.

    • If all allocation is used, users can run jobs in this partition only if they are associated with labs that have core hours in their Slurm lab banks. Usually, Slurm lab bank is a PI lab account.

1.5.1. Using interactive job

srun

To request an interactive job, use the srun command. Interactive does not mean instantaneous, the scheduler still needs to check for available resources and balance users and jobs.

To start an interactive session

You can use one of these example methods :

  1. Run job in standard partition, your default Slurm bank account will be charged.

    [user@login-x:~]$ srun --pty /bin/bash -i
srun: job 37382052 queued and waiting for resources
srun: job 37382052 has been allocated resources
[user@hpc3-l18-04:~]$

  2. Run job in free partition, your default Slurm bank account will not be charged.

    [user@login-x:~]$ srun -p free --pty /bin/bash -i
srun: job 37382053 queued and waiting for resources
srun: job 37382053 has been allocated resources
[user@hpc3-16-04:~]$

  3. Run job in standard partition using panteater_lab account (you must have an access to it). Your Slurm lab account will be charged.

    [user@login-x:~]$ srun -A panteater_lab --pty /bin/bash -i
srun: job 37382054 queued and waiting for resources
srun: job 37382054 has been allocated resources
[user@hpc3-19-01:~]$

After you execute the command, Slurm will put you on a compute node (when it is available) and you will see a new shell prompt in the terminal window. Now you can interactively run your applications and commands from the command line.

To stop an interactive session

After you are done, logout from interactive node. This will end your Slurm interactive session and you will return to the terminal window on the login node:

[user@hpc3-l18-04:~]$ logout
[user@login-x:~]$

1.5.2. Using batch job

sbatch

A batch job is run by the scheduler at sometime in the future and the scheduler picks an available time and node. Usually, it is within minutes, or as soon as requested resources become available. Slurm balances resource usage among many users and many jobs. If you are requesting resources that are not readily available you may have to wait longer.

Slurm batch jobs can be submitted to the same partitions as interactive jobs. A user needs to use:

  • sbatch command

  • Slurm submit script which is a text file that specifies:

    • the directives to Slurm what resources are needed for the job

    • commands that the jobs needs to execute

Example sbatch job

In the steps below you will download an example Slurm submit script, Python example program, then submit a job to the scheduler and check its output. All commands are executed on the cluster after you login.

Step: download example files

Files are downloaded from our web server to your user area that is allocated to you on the cluster. The submit script and Python program don’t need any editing after the download. Type all 4 commands exactly as they are shown (text after the prompt).

  1. Change directory to your DFS user area, here $USER is a shortcut for your UCInetID.

    [user@login-x:~]$ cd /pub/$USER

  2. Download Slurm submit script and save it as firstjob.sub file

    [user@login-x:~]$ wget https://rcic.uci.edu/_static/examples/firstjob.sub

  3. Download Python code and save it as days.py file. It is a simple Python program that prints today’s date and a random day 1-365 days in the past.

    [user@login-x:~]$ wget https://rcic.uci.edu/_static/examples/days.py

  4. Show the content of the Slurm script in the Terminal window.

    [user@login-x:~]$ cat firstjob.sub
Step: submit job to Slurm scheduler
[user@login-x:~]$ sbatch firstjob.sub
Submitted batch job 5776081

The output shows that script was submitted as a job with ID 5776081. All job IDs are unique, yours will be different and the output file name of your job will reflect a different ID.

Step: Check the job status and output file

This test job will run very quickly (fraction of a second) because it executes a few very fast commands and has no computational component.

Check the status of your job.

[user@login-x:~]$ squeue -u $USER
JOBID   PARTITION   NAME  USER  ACCOUNT ST   TIME  CPUS NODE NODELIST(REASON)

When the output is a single line as shown, the job is finished, otherwise there would be info about your job in the output.

List the files in the current directory.

[user@login-x:~]$ ls
firstjob.5776081.err  firstjob.5776081.out  firstjob.sub

There are now error and output files produced by the Slurm job as was requested in the submit script. Your file names will reflect your Slurm job ID.

Show the contents of the output file in the Terminal window.

[user@login-x:~]$ cat firstjob.5776081.out
Running job on host hpc3-l18-05
Today is 2021-07-23 and 325 days ago it was 2020-09-01

Note

Alwya check your slurm output and error files. They provide information about your job.

1.6. Storage

The filesystem storage is generally in 3 areas. Please use the links below for detailed information about each filesystem.

Important

All filesystems have storage quotas. Check quotas on regular basis after adding or removing a lot of files, transferring data or running computational jobs that create a lot of files:

HOME
All users have 50GB quota $HOME area. The $HOME is in /data/homezvolX/UCInetID. Use it for storing important and rarely changed files.
How to check $HOME quotas
DFS

  • All users have 1Tb quota /pub/UCInetID private area. Use it for storing data sets, documents, Slurm scripts and jobs input/output.

  • Depending on a lab affiliation, some users may have space in additional DFS areas (/dfs7, etc). This area is accessible by all members of the lab.

How to check DFS quotas

CRSP
By default users don’t have access to this area. Depending on a lab affiliation, some users may have space in /share/crsp/lab/labname/UCInetID. Please see Getting CRSP Account for details.
How to check CRSP quotas

1.7. Transferring files

Often users need to brings data from other servers and laptops. To transfer data one needs to use scp (secure copy) or rsync (file copying tool). Please see detailed Data transfer examples.

Alternatively, one can use graphical tools on their laptops (FileZilla, MountainDuck, or WinSCP) to transfer files between a local laptop and the cluster. Please follow each program instructions how to do this.

In all of the transfer application you will need to use hpc3.rcic.uci.edu as a remote server (where you want to transfer your files to/from) and use your UCInetID credentials for your user name and password.

Simple examples of file transfers with scp:

The scp command is used to transfer files and directories between a local laptop and a remote server. The command has a simple structure:

scp OPTIONS SOURCE DESTINATION
We omit OPTIONS for the simple cases.
The SOURCE and DESTINATION may be specified as a local file name, or a remote host with optional path in the form user@server:path where
user is your account on a cluster
@server: is the server name delimited with 2 special characters,
@ separates user name from server name
: separates server name from path name
path is a file path name on the server, can be made explicit using absolute or relative names:
absolute or full file name /Users/someuser/project1/input/my.fasta on your laptop
relative file name is my.fasta when used from the directory where this file is located.

  1. Transfer a single file to HPC3

    Transfer file myfile.txt from a local laptop to HPC3 and put it in the directory /pub/panteater. On your laptop, use a Terminal app and descend into the directory where your file is located, then execute the scp command (use your UCInetID):

    scp myfile.txt panteater@hpc3.rcic.uci.edu:/pub/panteater

  2. Transfer a single file from HPC3 to a local laptop

    On your laptop, use a Terminal app and descend into the directory where you want to put the transferred file, then execute the scp command (use your UCInetID):

    scp panteater@hpc3.rcic.uci.edu:/pub/panteater/project1/j-123.fa j-123.fa

  3. Transfer multiple files from your laptop to HPC3:

    scp f1.py f2.py doc.txt panteater@hpc3.rcic.uci.edu:/pub/panteater

  4. Transfer a directory with all its content from HPC3 to a local laptop

    On your laptop, use a Terminal app and descend into the directory where you want to put the transferred directory and its content, then execute the scp command (use your UCInetID). Note, the single dot character at the end means copy to this current directory:

    scp -r panteater@hpc3.rcic.uci.edu:/pub/panteater/results .

1.8. Logout

You can run many commands and submit many jobs. After you are done with your work you need to logout from the cluster using logout or exit command, for example:

[user@login-x:~]$ logout