High Performance Computing

High performance computing (HPC) is a necessary component of modern astrophysics research.  MKI has used 3 HPC clusters over the past decade.  The first cluster has been used by the Chandra space telescope for data analysis and modeling associated with the HETG instrument contract.  It was retired in October 2013.

The second cluster was used by the LIGO project, and was recently retired.  As LIGO undergoes upgrades to Advanced LIGO sensitivity, the computing and analysis infrastructure will be completely overhauled.

The third cluster was originally built by Professors Edmund Bertschinger and Scott Hughes, and largely used by their research groups.  It was significantly upgraded in 2008, replacing the original 32-bit compute nodes with 64-bit nodes.  It has been further upgraded to use CentOS Enterprise Linux with the Rocks Cluster Suites, an open source cluster management package.  The HPC cluster currently consists of 51 compute nodes and one master node, plus 4 storage servers providing researchers 1,048 CPU processors (cores) with 2,832 GB memory and 900 TB data storage.  We also have a Graphics Processing Unit (GPU) server providing nVidia's Tesla 2050 and 2075 cards with 896 GPU cores and 60 GB memory.

Over the past several years, this cluster has become a major computing resource for many MKI research groups.  The following groups have used the cluster for a variety of MKI research projects:

Scott Hughes Group
Modeling of gravitational-wave sources and their measurement by gravitational-wave detectors.  Modeling sources requires solving the equations of generic relativity to high accuracy, which requires large, parallel, high-precision HPC simulations.  Modeling gravitational-wave measurement requires large Monte-Carlo simulations in order to explore the parameter space of signals that nature might provide and to understand how well these signals can be distinguished from one another.

Edmund Bertschinger Group
Evolving with high precision and few approximations a system of N particles that interact through gravity.

George Clark
Computing x-ray scattering from dust in the interstellar medium (ISM).

TESS Group
Computing exoplanet atmosphere models.

Leslie Rogers (thesis project)
Modeling the interior structure of super-Earth and sub-Neptune exoplanets, as a contribution to the Kepler Science Group

Anna Frebel Group
Computing models of galactic haloes, and simulating galaxy formation. Galaxy formation is a complicated nonlinear process.  Researchers run large-scale N-body simulations to compare their theoretically predicted galaxies to observed ones.  Because these programs are computationally expensive and require lots of memory, Frebel's group uses the MKI HPC cluster to run and analyze these N-body simulations in order to gain intuition about the formation of galaxies.

Simona Vegetti  (former MKI postdoctoral fellow)
Modeling a large sample of gravitational lens galaxies.  The goal of this modeling is to determine the posterior probability density distribution of lens parameters, as well as the Bayesian evidence.  This allows Vegetti to objectively compare models that allow for the presence of mass clumps in the lens galaxy versus models that assume a smooth mass distribution.

MWA Group
Modeling the complex polarized beam patterns of the MWA tiles for their different pointing directions.