eResearch Doc

User Tools

Site Tools

You are here: start » hpc » faq
hpc:faq

Differences

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

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
hpc:faq [2025/06/04 08:29] – [Connection to Cluster] Yann Sagonhpc:faq [2025/09/01 07:39] (current) – Yann Sagon
Line 27: Line 27:
 !!!There could be several reasons for the cluster to slow down. It’s important to figure out where the slowness is happening: !!!There could be several reasons for the cluster to slow down. It’s important to figure out where the slowness is happening:
  
-  * **Login Node**:If the login node feels slow, it might be because someone is running heavy processes on it, which isn’t recommended. The login node is meant for tasks like file editing, job submission, and monitoring—not running jobs. If another user is hogging the CPU resources, it could affect your experience, but this won’t impact the performance of jobs on the compute nodes.+  * **Login Node**:The login node is designed for light tasks such as file editing, job submission, and monitoring, not for running heavy computations. To ensure fair usage and maintain responsiveness, each user is limited to 2 CPU cores and 8 GB of RAM on the login node
  
   * **Compute Nodes**: Slowness on the compute nodes might be due to high CPU usage, storage issues, or other factors, which could cause your jobs to run more slowly.   * **Compute Nodes**: Slowness on the compute nodes might be due to high CPU usage, storage issues, or other factors, which could cause your jobs to run more slowly.
Line 89: Line 89:
 </code> </code>
  
 +??? If multiple PIs jointly purchase computing resources on the Baobab cluster, who receives the invoice?
 +
 +!!! When several Principal Investigators (PIs) collaborate to acquire computing resources on the Baobab cluster, the invoice will be sent to the designated contact person of the group associated with the partition named private_xxx, where xxx is the name of the partition. This person acts as the reference for the group and is responsible for managing the billing and communication related to the shared resources.
 +
 +It is up to the PIs involved to agree among themselves on how the computing hours are distributed. The Baobab team does not manage internal allocations within shared purchases.
  
 ??? I'm a PI, I tried to use OpenXDmoD to see the past usage of my group without success ??? I'm a PI, I tried to use OpenXDmoD to see the past usage of my group without success
Line 94: Line 99:
  
  
-??? How can I check usage on more than one partition?+??? With OpenXDmoD how can I check usage on more than one partition?
 !!! Unfortunately, it seems that you need to do this operation for each partition separately. !!! Unfortunately, it seems that you need to do this operation for each partition separately.
  
Line 106: Line 111:
  
  
-??? I'm organising a course and we need some HPC resources for the students. Do we have to pay for it?+??? I'm organizing a course and we need some HPC resources for the students. Do we have to pay for it?
 !!! The Baobab service is free for courses as long as the usage is low and for a defined period of time. Check [[hpc:hpc_clusters#use_baobab_for_teaching|How our clusters work]]. !!! The Baobab service is free for courses as long as the usage is low and for a defined period of time. Check [[hpc:hpc_clusters#use_baobab_for_teaching|How our clusters work]].
  
Line 125: Line 130:
 !!!  * If you have a non student account (Phd, postdoc, researcher), your account will expire at the same time your contract expire at UNIGE. Right now, there is a grace period after the end of your contract of around 6 months. !!!  * If you have a non student account (Phd, postdoc, researcher), your account will expire at the same time your contract expire at UNIGE. Right now, there is a grace period after the end of your contract of around 6 months.
   * If you have an outsider account, you need to check the expiration date you received when you filled the invitation.   * If you have an outsider account, you need to check the expiration date you received when you filled the invitation.
-  * If you have an unige student account, you can check the expiration date with the ''chage'' command:+  * If you have an UNIGE student account, you can check the expiration date with the ''chage'' command:
 <code> <code>
 (baobab)-[yourusername@login2 ~]$ chage -l yourusername (baobab)-[yourusername@login2 ~]$ chage -l yourusername
Line 138: Line 143:
  
 ??? I'm leaving UNIGE, can I continue to use Baobab HPC service? ??? I'm leaving UNIGE, can I continue to use Baobab HPC service?
-!!! Yes it is possible as long as you collaborate tightly with your former research group. Your PI must [[https://gestion-externe.unige.ch/main/outsider-requests|invite]] you as [[hpc:access_the_hpc_clusters#outsider_account|outsider]]. For technical reason, your account needs to be expired  prior doing the request for the invitation. +!!! For UNIGE or external members, please first refer to the guidelines about account expiration and the grace period: https://plone.unige.ch/distic/pub/isis/comment-fonctionne-isis#prolongations. 
-We'll then reactivate your account. You'll keep your data.+Please note that expired accounts are eligible for deletion at any time. We strongly recommend that you carefully prepare for your departure or contract extension.
  
 +However, it is possible to extend your access as long as you maintain close collaboration with your former research group. Your PI must [[https://gestion-externe.unige.ch/main/outsider-requests|invite]] you as an [[hpc:access_the_hpc_clusters#outsider_account|outsider]].
 +For technical reasons, your account must be expired before making the invitation request.
 +Once the invitation is processed, we will reactivate your account, and you will retain access to your data.
  
  
  
  
-=?=== Connection to Cluster ===+ 
 +=?=== Connection to Cluster ====
  
 ??? When I type my password, no characters are printed. Why? ??? When I type my password, no characters are printed. Why?
Line 216: Line 225:
  
  
-=?==== X2GO-Desktop =====+=?=== X2GO-Desktop ====
 ??? Why I can't connect with x2go ? ??? Why I can't connect with x2go ?
 !!! We have already identified a number of common problems: !!! We have already identified a number of common problems:
Line 245: Line 254:
     - **~/.local/session**     - **~/.local/session**
     - **~/.config/xfce**     - **~/.config/xfce**
-=?==== Storage =====+=?=== Storage ====
  
 ??? I have a question about the storage !? ??? I have a question about the storage !?
Line 267: Line 276:
 !!! If you have a lot of data, the best way is to use rsync between both clusers, so you won't have to copy the data to your laptop first. [[hpc:best_practices#transfer_data_from_one_cluster_to_another|Transfer data from one cluster to another]] !!! If you have a lot of data, the best way is to use rsync between both clusers, so you won't have to copy the data to your laptop first. [[hpc:best_practices#transfer_data_from_one_cluster_to_another|Transfer data from one cluster to another]]
  
-=?==== Applications =====+=?=== Applications ====
  
 ??? What applications are installed on Clusters ?  ??? What applications are installed on Clusters ? 
Line 328: Line 337:
  
 In this case you need to check if there is another version available compatible with the toolchain (''GCC'', ''foss'' etc...) you want to use. If not, please refer to [[hpc:faq#the_software_i_need_is_not_ava|The software I need is not available on Clusters: what should I do ?]].  In this case you need to check if there is another version available compatible with the toolchain (''GCC'', ''foss'' etc...) you want to use. If not, please refer to [[hpc:faq#the_software_i_need_is_not_ava|The software I need is not available on Clusters: what should I do ?]]. 
-=?==== Slurm: job scheduler =====+=?=== Slurm: job scheduler ====
 ??? What is Slurm ? ??? What is Slurm ?
 !!! Slurm is a job scheduling system used to manage and allocate resources in a computing cluster. It helps you submit, monitor, and control jobs (tasks) on the cluster. !!! Slurm is a job scheduling system used to manage and allocate resources in a computing cluster. It helps you submit, monitor, and control jobs (tasks) on the cluster.
Line 410: Line 419:
  
 Please send an email to [[hpc@unige.ch]] including relevant information (Uusername, Group, private_partion etc...) with the responsible person for the share or partition in CC. The responsible person **must** approve the modification. Please send an email to [[hpc@unige.ch]] including relevant information (Uusername, Group, private_partion etc...) with the responsible person for the share or partition in CC. The responsible person **must** approve the modification.
-=?==== Mac Issues  =====+ 
 +=?=== Issues  ====
  
 ??? I have a keyboard issue using a Mac. ??? I have a keyboard issue using a Mac.
Line 431: Line 441:
 Refer to this solution: [[https://stackoverflow.com/questions/50035949/macos-high-sierra-and-x11-forwarding/50182736#50182736|macOS High Sierra and X11 Forwarding]]. Refer to this solution: [[https://stackoverflow.com/questions/50035949/macos-high-sierra-and-x11-forwarding/50182736#50182736|macOS High Sierra and X11 Forwarding]].
  
-=?==== Switch edu-ID Login Issues =====+=?=== Switch edu-ID Login Issues ====
  
 ??? I get an error message from Switch edu-ID while trying to access: ??? I get an error message from Switch edu-ID while trying to access:
Line 446: Line 456:
 Please also note that your ISIS (UNIGE) password and your Switch edu-ID password are not the same. Verify that you are using the correct password when logging in. Please also note that your ISIS (UNIGE) password and your Switch edu-ID password are not the same. Verify that you are using the correct password when logging in.
  
-=?==== HPC community forum =====+=?=== HPC community forum ====
  
 ??? I don't find a way to receive email summary of new post ??? I don't find a way to receive email summary of new post
hpc/faq.1749025788.txt.gz · Last modified: (external edit)