Hi Everyone,
this is only short reminder, that there is MedBioNode Space available on Element, internal MedUni Graz communication platform. You can get part over the link:
https://matrix.to/#/#medbionode:medunigraz.at
and post relevant staff around our HPC cluster.
Best regards,
Slave
--
Slave Trajanoski, Phd
Senior Scientist Bioinformatics
CF Computational Bioanalytics, Center for medical research
Medical University Graz
Neue Stiftingtalstraße 6 - West Tower P 4th Floor
8010 Graz
Tel. +43 316 385 73024
E-Mail: slave.trajanoski(a)medunigraz.at<mailto:slave.trajanoski@medunigraz.at>
Hi everyone,
Many slurm jobs are running for days, and some of them have been going for almost a week. On the other hand, we have a lot of jobs that need to be done but are waiting for free cluster resources. I strongly advise that you do not start a lot of long-running slurm jobs simultaneously. This will block most of the cluster's resources, making them unavailable to other users. Instead, I suggest that you reduce the number of jobs you start or split them up into smaller pieces if you can. We're following the fair use principle and trying to stay away from rules and restrictions, which often make things more difficult to understand and use, as well as less flexible and efficient. Please take a look at your active jobs today and see where you can trim or pause them so your colleagues can get some work done.
Thanks for your help.
Best regards,
Slave
--
Slave Trajanoski, Phd
Senior Scientist Bioinformatics
CF Computational Bioanalytics, Center for medical research
Medical University Graz
Neue Stiftingtalstraße 6 - West Tower P 4th Floor
8010 Graz
Tel. +43 316 385 73024
E-Mail: slave.trajanoski(a)medunigraz.at<mailto:slave.trajanoski@medunigraz.at>
Dear RStudio Server users,
please use the following recommendations for smooth work with RStudio Server, as well as avoiding potential issue for you or possibly causing disturbance to other users too on our MedBioNode cluster. Thanks very much in advance.
To ensure the login node remains stable and responsive for everyone, please follow these guidelines when using the Open Source RStudio Server web interface.
1. Always use "Sign Out" — Don't just close the tab
* The Problem: If you close your browser tab or laptop lid without logging out, your R session stays running in the background on the HPC login node. If you reconnect later from a different network, device, or browser, a duplicate session is created. This wastes system memory and slows down the cluster.
* The Solution: When you are done working for the day, click the Sign Out button in the top-right corner of the RStudio window. This safely terminates your background processes.
2. Clean up your environment before leaving
Before signing out, it is highly recommended to clear your R workspace to prevent bloated session files that slow down your next login:
* Run rm(list = ls()) in your console to clear memory.
* Go to Tools -> Global Options -> General and change “Save workspace to .RData on exit” to Never. This stops RStudio from saving massive, fragmented files to your home directory, which can cause your next session to hang on startup.
Best regards,
Slave
--
Slave Trajanoski, Phd
Senior Scientist Bioinformatics
CF Computational Bioanalytics, Center for medical research
Medical University Graz
Neue Stiftingtalstraße 6 – West Tower P 4th Floor
8010 Graz
Tel. +43 316 385 73024
E-Mail: slave.trajanoski(a)medunigraz.at<mailto:slave.trajanoski@medunigraz.at>
Dear MedBioNode users,
login node from our MedBioNode cluster (medbionode.medunigraz.at or sx163) will be shut down in 60 minutes for maintenance purpose. Please save your work.
Thanks for your understanding.
Best regards,
Slave
--
Slave Trajanoski, Phd
Senior Scientist Bioinformatics
CF Computatinoal Bioanalyitcs, Center for medical research
Medical University Graz
Neue Stiftingtalstraße 6 - West Tower P 4th Floor
8010 Graz
Tel. +43 316 385 73024
E-Mail: slave.trajanoski(a)medunigraz.at<mailto:slave.trajanoski@medunigraz.at>
Dear all,
we have identified issues with JupyterHub running on the login node sx163, caused by stalled user processes that cannot be stopped normally. To resolve this and restore full service, it will be necessary to restart the server.
Impact:
The JupyterHub service and RStudio server on sx163 will be temporarily unavailable during the restart.
Any currently running Jupyter sessions may be interrupted. Please save your work before the restart if possible.
We apologize for the inconvenience and appreciate your understanding. The server will be restarted in 20 minutes to minimize downtime.
Thank you for your cooperation,
Slave
--
Slave Trajanoski, Phd
Senior Scientist Bioinformatics
CF Computatinoal Bioanalyitcs, Center for medical research
Medical University Graz
Neue Stiftingtalstraße 6 - West Tower P 4th Floor
8010 Graz
Tel. +43 316 385 73024
E-Mail: slave.trajanoski(a)medunigraz.at<mailto:slave.trajanoski@medunigraz.at>
Dear MedBioNode users,
I'd like to announce that since today we have additional GPU node with four Nvidia L40s GPUs.
Regards,
Slave
--
Slave Trajanoski, Phd
Senior Scientist Bioinformatics
CF Computatinoal Bioanalyitcs, Center for medical research
Medical University Graz
Neue Stiftingtalstraße 6 - West Tower P 4th Floor
8010 Graz
Tel. +43 316 385 73024
E-Mail: slave.trajanoski(a)medunigraz.at<mailto:slave.trajanoski@medunigraz.at>
Dear All,
I am writing to share information about two recently opened positions in the fields of Bioinformatics at our Core Facility for Computational Bioanalytics. Detailed descriptions of these opportunities are provided in the attached document.
If you know of any colleagues or contacts who might be interested, please feel free to forward this information to them.
Thank you for helping us spread the word!
Best regards,
Slave
--
Slave Trajanoski, Phd
Senior Scientist Bioinformatics
CF Computatinoal Bioanalyitcs, Center for medical research
Medical University Graz
Neue Stiftingtalstraße 6 - West Tower P 4th Floor
8010 Graz
Tel. +43 316 385 73024
E-Mail: slave.trajanoski(a)medunigraz.at<mailto:slave.trajanoski@medunigraz.at>
Dear MedBioNode users,
I have great news for you. IT department managed to issue new certificates for the two critical services much faster than expected:
1. jupyter.medunigraz.at
2. rstudio.medunigraz.at
which are running on the new login node sx163. You can access them with any browser over https secure connection. Many thanks to the responsible person at the IT department. The new server has much better and faster hardware: 256 logical cores and 1.5TB working memory. Herewith you can also ignore my previous E-Mail.
Best regards,
Slave
--
Slave Trajanoski, Phd
Senior Scientist Bioinformatics
CF Computational Bioanalytics, Center for medical research
Medical University Graz
Neue Stiftingtalstraße 6-West, Turm/Stiege:P, 4.Stock
Tel. +43 316 385 73024
E-Mail: slave.trajanoski(a)medunigraz.at<mailto:slave.trajanoski@medunigraz.at>
Dear MedBioNode users,
I have good news. We're providing additional login node, RStudio Server and JupyterHub over the following links:
1. Second login node ssh sx163.medunigraz.at
2. RStudio server http://sx163.medunigraz.at:8787
3. JupyterHub https://sx163.medunigraz.at
For RStudio server and JupyterHub you'll need to use Firefox and make exception for this address until IT department provides certificates for them.
Since both services will run in your one and same user profile you must avoid opening sessions at the same time on current servers (jupyterhub.medunigraz.at and rstudio.medunigraz.at) and the new ones.
If you have doubts, need support or just ask questions don't hesitate to contact me.
Best regards,
Slave
--
Slave Trajanoski, Phd
Senior Scientist Bioinformatics
CF Computational Bioanalytics, Center for medical research
Medical University Graz
Neue Stiftingtalstraße 6-West, Turm/Stiege:P, 4.Stock
8010 Graz
Tel. +43 316 385 73024
E-Mail: slave.trajanoski(a)medunigraz.at<mailto:slave.trajanoski@medunigraz.at>
Dear MedBioNode users,
herewith we’re introducing new way for encrypted communication using save local server installed and administered by the IT departement. This is the documentation on how to install and configure your chat client:
https://confluence.medunigraz.at/display/ITKB/Matrix+%28Element%29+Messenger
and here the invitation for the MedBioNode space.
https://matrix.to/#/#medbionode:medunigraz.at
You can contact me also directly over this channel. The already existing Discord channel will not and should not be used in the future.
Best regards,
Slave
--
Slave Trajanoski, Phd
Senior Scientist Bioinformatics
CF Computational Bioanalytics, Center for medical research
Medical University Graz
Neue Stiftingtalstraße 6-West, Turm/Stiege:P, 4.Stock
Tel. +43 316 385 73024
E-Mail: slave.trajanoski(a)medunigraz.at<mailto:slave.trajanoski@medunigraz.at>