IT wiki

VU MIF STSC

User Tools

Site Tools


en:hpc:getting_started

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

en:hpc:getting_started [2026/11/19 13:48] – created eduardasen:hpc:getting_started [2026/11/19 13:58] (current) – removed eduardas
Line 1: Line 1:
-====== HPC Quick Start (15 Minutes) ====== 
  
-This guide walks you through the basic HPC workflow: 
- 
-  - Connect to the HPC system 
-  - Create a simple job script 
-  - Submit your first job 
-  - View the results 
- 
-If you can complete this guide, you are ready to use the HPC cluster for your own workloads. 
- 
-===== Before You Start ===== 
- 
-Make sure that: 
- 
-  * You have access to the HPC system. 
-  * You have completed [[hpc:registration|Registration]]. 
-  * You have an SSH client installed. 
- 
-Need access first? 
- 
-  * [[hpc:registration|Register for HPC Access]] 
- 
-===== Step 1: Connect to the Cluster ===== 
- 
-Open a terminal and connect to the login node: 
- 
-<code bash> 
-ssh username@login.hpc.mif.vu.lt 
-</code> 
- 
-Replace ''username'' with your HPC username. 
- 
-After logging in, verify your current location: 
- 
-<code bash> 
-pwd 
-</code> 
- 
-You should be in your home directory. 
- 
-For detailed connection instructions see: 
- 
-  * [[hpc:connection|Connection]] 
- 
-===== Step 2: Create a Working Directory ===== 
- 
-Create a directory for this tutorial: 
- 
-<code bash> 
-mkdir getting-started 
-cd getting-started 
-</code> 
- 
-Verify your location: 
- 
-<code bash> 
-pwd 
-</code> 
- 
-===== Step 3: Create a Simple Job Script ===== 
- 
-Create a file named ''hello.sh'': 
- 
-<code bash> 
-nano hello.sh 
-</code> 
- 
-Paste the following content: 
- 
-<code bash> 
-#!/bin/bash 
-#SBATCH --job-name=hello 
-#SBATCH --output=hello.out 
-#SBATCH --time=00:01:00 
- 
-echo "Hello from HPC!" 
-hostname 
-date 
-</code> 
- 
-Save the file. 
- 
-Make it executable: 
- 
-<code bash> 
-chmod +x hello.sh 
-</code> 
- 
-===== Step 4: Submit Your First Job ===== 
- 
-Submit the job to SLURM: 
- 
-<code bash> 
-sbatch hello.sh 
-</code> 
- 
-Example output: 
- 
-<code> 
-Submitted batch job 12345 
-</code> 
- 
-The number is your Job ID. 
- 
-===== Step 5: Check Job Status ===== 
- 
-View your jobs: 
- 
-<code bash> 
-squeue -u $USER 
-</code> 
- 
-When the job disappears from the queue, it has finished. 
- 
-===== Step 6: View the Results ===== 
- 
-Display the output file: 
- 
-<code bash> 
-cat hello.out 
-</code> 
- 
-Example output: 
- 
-<code> 
-Hello from HPC! 
-compute01 
-Mon Nov 17 14:32:05 EET 2026 
-</code> 
- 
-Congratulations! You have successfully executed your first HPC job. 
- 
-===== Useful Commands ===== 
- 
-^ Command ^ Purpose ^ 
-| ''pwd'' | Show current directory | 
-| ''ls'' | List files | 
-| ''cd'' | Change directory | 
-| ''sbatch job.sh'' | Submit a job | 
-| ''squeue -u $USER'' | View your jobs | 
-| ''scancel JOBID'' | Cancel a job | 
-| ''cat file'' | Display file contents | 
- 
-===== Next Steps ===== 
- 
-Now that you can run jobs, learn how to: 
- 
-  * [[hpc:lustre|Store and manage data using Lustre]] 
-  * [[hpc:slurm|Request resources with SLURM]] 
-  * [[hpc:resources|Understand quotas and project allocations]] 
- 
-===== Common Problems ===== 
- 
-==== Permission denied ==== 
- 
-Verify that your account has been activated and that you are using the correct username. 
- 
-==== sbatch: command not found ==== 
- 
-Ensure you are connected to the HPC login node and not running commands locally. 
- 
-==== Invalid account or account/partition combination specified ==== 
- 
-Your SLURM account may still be provisioning. Wait a few minutes after your first login and try again. 
- 
-==== Job stays in PENDING state ==== 
- 
-Resources may currently be unavailable. Check the reason: 
- 
-<code bash> 
-squeue -j JOBID 
-</code> 
- 
-For more information see: 
- 
-  * [[hpc:slurm|Batch Job Execution (SLURM)]] 
- 
-===== Where to Go Next ===== 
- 
-Recommended learning path: 
- 
-  - [[hpc:registration|Registration]] 
-  - [[hpc:getting_started|Quick Start]] 
-  - [[hpc:connection|Connection]] 
-  - [[hpc:lustre|Shared File System (Lustre)]] 
-  - [[hpc:slurm|Batch Job Execution (SLURM)]] 
-  - [[hpc:resources|Resource Allocations]] 
-  - [[hpc:waldur|Waldur User Guide]] 

Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki