Search Results

The latest Lustre documentation is available at Lustre

Distributed filesystems such as Lustre are ideal for HPC and HTC environments. In these environments the typical workload consists of large files that have to be accessed from many compute nodes with very high bandwidth and/or low latency. Therefore these file...

Public login services Access to the CNCA clusters is only available to registered users. To use the HPC services users need to login via SSH into a front-end server. The login servers can be customized according to the user requirements (OS, Hardware resource...

Access to the CNCA computing clusters is performed via SSH and requires the use of SSH keys for authentication. Authentication with passwords is not supported. Each SSH key pair has two components a public key that must be added to the hosts to be remotely acc...

Choose strong passwords - at least 9 characters long, a mixture of alphanumeric mixed case and symbol characters. The password should be completely different from the password you use on any other system. Never use the same password across different systems ...

In order to simplify SSH access to remote hosts we recommend CNCA users to adopt the following recommendatios and configuration defaults. Enabling SSH agents and X11 forwarding Activate the forwarding of both SSH authentication credentials and X11 graphical wi...

CNCA is open to all academic and scientific researchers and institutions. In order to use our services please go to the CNCA portal and follow the instructions there. For guidance and further information about the access conditions see the FAQ List warning Pl...

Access to the CNCA cloud and distributed computing infrastructure is available only to registered users. Access to login machines may be filtered by IP address of origin. During the registration process you may be asked to provide your IP address in order to ...

At Cirrus the default max elapsed time for the queues is 72h. The values for all queues can be consulted here In case you cannot parallelize your job (split one single job into multiple parallel jobs) you can use job dependencies. In this case you create ...

The Cirrus filesystems are based on the Lustre shared filesystem which is mounted in the compute nodes and in the submission/login nodes. Lustre is mostly suited to the storage of large/huge files. Summary CNCA-Lisbon cluster Name[1] Purpose Availability Qu...

To prevent storage misuse and problems resulting from filesystems becoming full, Cirrus has implemented the following data usage policies and limits. File system Hard Limit Exceeded /home Once quota or number of files exceeded, new files will not be all...

NOTE Please note that the Cirrus Lustre storage is not designed for small files. Attempting to store or retrieve files less than a few megabytes will result in extremely poor performance for all users. If you wish to store lots of small files to massdata, ple...

Users are allowed to change the default permissions file permissions in all folders they own. The default access rights to Cirrus filesystem is the folllowing: /home All files under /home are only accessible by the user. /data All files under /data/unixgrp/...

Small tutorial on how to install and run miniconda on the HPC clusters. 1. Login into your login machine 2. Download Miniconda into your local home $ wget https://repo.anaconda.com/miniconda/Miniconda2-latest-Linux-x86_64.sh 3. Execute Miniconda $ ch...

The Cirrus software is managed using a tool called modules. This tool allows the user to load the correct environment (PATH, LD_LIBRARY_PATH, etc) for each specific software package. The main commands are explained in the next sections. List of available Softw...

There are a few options for local users installation locations as showed on the next table, create appropriate paths bellow the base directory. Site Base Directory Comments CNCA-Lisbon (Cirrus-A) /home/GROUP/USERNAME CNCA-Lisbon (Cirrus-A) /data/GROU...

For specific user portal issues please use the following contact: portal (at) acnca.pt

List of Queues INCD-Lisbon cluster (cirrus.a.incd.pt) Name Jobs max elapsed time access Memory Max #cores[1] Comments gpu 72h everyone 2 GB/Core 16 queue for GPU resources hpc 72h everyone 8 GB/Core 64 default queue fct 72h reser...

CNCA is required to provide evidence of the work made possible through the usage of the infrastructure services. The metric of success is the scientific publications produced through the usage of the CNCA services. Please help us to continue delivering service...

Journal Paper with direct contribution Quarkonium production at the LHC: A data-driven analysis of remarkably simple experimental patterns P. Faccioli, C. Lourenço, M. Araújo, V. Knünz, I. Krätschmer, J. Seixas Phys.Lett. B773 (2017) 476-486 10.1016/j.physle...