NYU's High Performance Computer Cluster

A Toolbox Supporting a Range of
Research Applications

Joseph Hargitai

Enhancing the NYU General Cluster’s "toolbox" are a number of intriguing computational tools added particularly to support research and instruction on the cluster over the past year. For example a new compiler from Berkeley Labs, Unified Parallel C (UPC, upc.lbl.gov), was used for class instruction in parallel computing by Professors Marsha Berger and David Bindel (Computer Science, CIMS). Luca Maragliano deployed a modified version of DLPROTEIN (www.sissa.it/cm/DLPROTEIN), a molecular dynamic software package for macromolecules. Roman Scoccimarro used Gadget II (www.mpa-garching.mpg.de/gadget), a code for cosmological N-body/SPH simulations on massively parallel computers with distributed memory. Daniel Goldberg used mathematical libraries, deal.II, A Finite Element Differential Equations Analysis Library, and PETSc, Portable, Extensible Toolkit for Scientific Computation. Finally, one of the popular "new" tools has actually turned out to be FORTRAN! Roine Vestman introduced the use of this venerable compiler to speed up his application, formerly scripted in Matlab. Here are a few of the other tools available on the cluster:

ROOT is a data processing framework from CERN, at the heart of the research on high-energy physics. Every day, thousands of physicists use ROOT applications to analyze their data or to perform simulations. root.cern.ch/drupal/faq

Migrate-n estimates effective population sizes and past migration rates between n populations assuming a migration matrix model with asymmetric migration rates and different subpopulation sizes. popgen.sc.fsu.edu/Migrate-n.html

FMS, Flexible Modeling System, is a software framework for supporting the efficient development, construction, execution, and scientific interpretation of atmospheric, oceanic, and climate system models. www.gfdl.noaa.gov/fms

Unified Parallel C (UPC) is an extension of the C programming language designed for high performance computing on large-scale parallel machines. The language provides a uniform programming model for both shared and distributed memory hardware. upc.lbl.gov

EEGLAB is an interactive Matlab toolbox for processing continuous and event-related EEG, MEG and other electrophysiological data incorporating independent component analysis (ICA), time/frequency analysis, artifact rejection, event-related statistics, and several useful modes of visualization of the averaged and single-trial data. sccn.ucsd.edu/eeglab

Structure is a free software package for using multi-locus genotype data to investigate population structure. Its uses include inferring the presence of distinct populations, assigning individuals to populations, studying hybrid zones, and identifying migrants and admixed individuals. pritch.bsd.uchicago.edu/structure.html

Rmpi is a package for R, a free software environment for statistical computing and graphics. It is an interface to MPI (Message-Passing Interface), a standardized and portable message-passing system designed to function on a wide variety of parallel computers.

Further Information About the Cluster

The NYU General Cluster offers 2.656 terabytes of combined memory, large-scale storage options, and a theoretical peak performance of 10.5 Teraflops; measured performance peaked at 8.418 Teraflops. The cluster consists of 140 compute nodes and two interactive login nodes, with a total CPU core count of 1120. Each compute node contains two Xeon Intel Quad-Core 64-bit processors, running at 2.33GHz. The cluster accommodates low-latency parallel runs via nodes that are connected with an Infiniband 4xDDR network from Cisco Systems, Inc.

Reservations for node segments, class use, special projects, and deadlines are available. For additional details, or for information on accessing the NYU General Cluster, please send email to hpc@nyu.edu.