UPDATE December 2025: We added another 12TB drive to the Planetary Materials Group server for storing the data for QME-Tool. The seemingly infinite 160TB drive on the seismology server keeps filling up so we have added an additional 150TB NAS, this time direct connected to the server with a 10Gbps connection. Our new faculty member, Dougal Hansen, has also purchased his own 100TB Synology NAS as a repository for his seismic data which is currently being processed using the seismology R7525 server. Slava Solomatov has also added two AMD Ryzen Threadripper 9955WX powered Precision 7875 workstations for his postdoc and graduate student.
In 2021 the seismology group added a PowerEdge R7525 running dual 3.3/4.0GHz 75F3 32 core processors and 512GB of RAM along with twelve 18TB hard drives in a 160TB RAID6 volume and a 3.8TB SSD for faster data processing. The addition of this additional server means we have 256 total AMD EPYC cores and 1.5TB of RAM available for use among our three R7525 servers, equivalent to about 10 of the 32 24-core nodes in the TELLUS cluster.
In September 2020 we installed two Dell PowerEdge R7525 computational servers, one for Bradley Jolliff’s Planetary Materials Group and one for Bronwen Konecky’s Climate and Paleoclimate Lab. The Planetary Materials server consists of dual 2.0/3.3GHz AMD EPYC 7662 64 core processors, 512GB of RAM and 24TB of local storage. The Paleoclimate server contains dual 2.9/3.4GHz AMD EPYC 7542 32 core processors, 512GB of RAM and a 25TB SSD RAID. It communicates with the 8-bay Paleoclimate Synology DiskStation NAS which acts as a 50TB data archive for the laboratory.
The Planetary Materials server primarily runs parallel Hapke photometric model optimizations of lunar images on MATLAB and the ISIS digital image processing package to analyze lunar images. The Paleoclimate server runs global paleo-climate model simulations as well as R (R-Studio/R-Server) and MATLAB. Both are running Ubuntu 22.04 LTS Linux and are available for faculty, staff and students to use for research projects within each laboratory.
If compiling your own code on these systems, the applications should be compiled using the AMD Optimizing C/C++ Compilers (AOCC). These have been installed on the systems and the user guide for using them can be found at the AOCC Home Page. The compiler itself is installed in /opt/AMD on the system so make sure to add /opt/AMD/aocc-compiler-3.0.0/bin to your path.