Resources

Computational Infrastructure

The core competency of the LAC is its ability to provide high-performance computing (HPC) resources tailored to the specific needs of astronomical research. Modern astrophysics often involves the resolution of large systems of differential equations, the processing of gigabyte-scale data cubes from integral field units (IFUs), and the execution of Monte Carlo simulations for statistical inference. To meet these demands, the LAC maintains a hardware environment that has been progressively upgraded through grants from agencies such as FAPEMIG, CNPq, and CAPES.

The Giskard Cluster

The centerpiece of the laboratory's computational power is the Giskard cluster. Giskard is a high-density computing system designed for parallel processing tasks. It consists of six calculation nodes, each powered by processors running at 2.20 GHz, providing a total of 240 CPUs across the nodes, plus a headnode with 4 CPUs at 1.87 GHz. The memory configuration of Giskard is particularly noteworthy, with more than 500 GB of total RAM, allowing for the handling of large datasets in-memory, which is crucial for reducing I/O bottlenecks in complex simulations. The physical housing of the cluster is a 25 $m^2$ climate-controlled room in the university datacenter, which ensures hardware longevity and operational stability in the face of environmental fluctuations.

Citation Policy

To properly acknowledge the use of our resources, all collaborators and researchers are required to include a formal citation in their publications. Recognizing the laboratory’s role is essential for maintaining the grants from CAPES, CNPq, and FAPEMIG that support our clusters and infrastructure.

Please use the following text in the acknowledgments section of your work:

"This work has made use of the computing facilities available at the Laboratory of Computational Astrophysics of the Universidade Federal de Itajubá (LAC-UNIFEI). The LAC-UNIFEI is maintained with grants from CAPES, CNPq and FAPEMIG." 

If you have additional questions regarding the citation protocol or need specific details about the hardware used for your simulations, please contact us directly at lac@unifei.edu.br.