Acquisition of Computing Facilities to support Collaborative Geophysics Research at the University of New Mexico

University Of New Mexico, Albuquerque NM

Investigators

Abstract

Research in modern geophysics increasingly requires access to significant computational resources which are needed to analyze large datasets using techniques such as InSAR time-series, as well as active and passive seismic imaging of the subsurface, and finite element modeling of subsurface processes. These analysis techniques require thousands of CPU-hours and can involve tens to hundreds of terabytes of data which are beyond the scale that workstations or small servers can manage. In addition, it is not practical to use cloud computing due to the need to move and store very large datasets. This award will significantly expand the capabilities of the University of New Mexico (UNM) in each of the research areas mentioned above through the acquisition of dedicated computational resources for storage and analysis of large geophysical datasets. The award also includes support for development of a Computational Earth Sciences summer workshop for undergraduates, led by the PIs and graduate students. By providing access to critical training in programming and high-performance computing, this workshop will encourage students from under-represented groups in STEM to continue a career in Earth sciences research and other computational fields. This project received co-funding from the Established Program to Stimulate Competitive Research (EPSCoR). This award will support the acquisition of a shared geophysics computing facility with two primary components: (1) a storage server with 192 TB capacity that will host geophysical datasets and processing results, and (2) a cluster of 4 dedicated computing nodes with 32 CPUs and 192 GB RAM each, that will run alongside the storage server. These resources will be hosted at the UNM Center for Advanced Research Computing (CARC), improving the ability to rapidly prototype code and store datasets larger than currently supported by existing CARC resources, while enabling access to larger-scale processing through CARC when needed. Together, these resources will enable us to quickly and seamlessly conduct large-scale computational projects that are presently beyond the reach of this institution. This computational resource will enable development of high-fidelity, data-driven models to investigate issues of particular concern to New Mexico, including: (1) the occurrence of injection-induced earthquakes in the Raton Basin, (2) magma storage and inflation of the Socorro Magma Body and the Valles Caldera, and (3) assessment of the long-term sustainability of actively exploited aquifers across the state. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.