Skip to Navigation | Skip to Content

High Performance & Computational Resources

search the site

Have a question about using high performance computing resources in support of your research? Send email to hpc@nyu.edu to request a consultation.

related pages

NYU supports high performance computing and networking for researchers and scholars, and is home to several high performance clusters and high-speed networks equipped with a wide variety of research software packages. If you or a member of your research group has high performance computing or networking needs—including visualization, simulations, or other data intensive operations, send email to hpc@nyu.edu to request a consultation. Use of ITS high performance computing resources is open to all faculty and to sponsored graduate students. ITS is also available to partner with NYU faculty as they seek funding for research with substantial technology components.

The NYU General Cluster

The new NYU General Cluster is a general purpose high performance computing resource for the NYU research community that offers 1.28 terabytes of memory, large-scale storage options, and a theoretical peak performance of more than 5 Teraflops (a trillion calculations per second). The Rocks operating system is used to manage the cluster, which consists of 64 compute nodes and two interactive login nodes, with a total CPU core count of 512. 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 DDR network from CISCO. The General Cluster offers GNU, Intel C, and Fortran compilers, and the Matlab, R, and Mathematica applications (additional software may be installed upon request). Home directories are mounted from the control node of the cluster, and there is also a scratch filesystem accessible from all nodes. For clients with large data needs, there is now a data filesystem attached to the clusters' login nodes. For additional details, or for information about how to access the NYU General Cluster, please send email to hpc@nyu.edu.

The SGI Cluster: GAUSS

GAUSS is a 64 CPU, 128Gb Altix 4700 server which runs SuSE Linux® and is based on Intel® Itanium® 2 dual core processors. The design of the server utilizes SGI® NUMAflex® architecture which permits computations, memory and I/O scaling within a single shared memory environment. This low-latency cluster provides exceptional performance.

The Max Supercomputer

a photograph of the max supercomputer

NYU's IBM eServer BladeCenter system is capable of a peak performance of 4.5 teraflops (one trillion floating point operations per second). The acquisition of this supercomputer, which supports NYU research with heavy computational requirements, was made possible by a gift from IBM and federal funding. The supercomputer has been dubbed "Max," in honor of Max Goldstein, a computing pioneer and—among his many accomplishments—the first director of NYU's Academic Computing Facility, one of Information Technology Services' predecessor organizations. See the Connect Magazine article links in the sidebar for additional information about Max.

Case Studies

The following case studies in research that utilizes NYU high performance computing resources were generously contributed by University researchers. If you have research that you would like to share with the NYU community, please send email to hpc@nyu.edu.

This video, produced by Yue-Kin Tsang, a post-doctoral student at NYU's Center for Atmosphere Ocean Science, shows the simulation of the mixing of a tracer (or "dye") by a circulation that is meant to represent the ocean. This project is concerned with understanding how to predict the mixing of that tracer on small scales by the kind of turbulence that takes place on a hundred-kilometer scale in the ocean. In the real ocean, climate scientists need to know how to predict the mixing of many kinds of tracers (heat, salt, plankton, etc...), but all obey essentially the same equations of motion.
This visualization, generated on Max by Ross Tulloch, a graduate student at NYU's Center for Atmosphere Ocean Science, shows a large scale evolution in time of a the temperature field in a particular kind of turbulence believed to be active at the tropopause (the upper boundary of the troposphere). As you can see, the filaments between the big vortices seem to "curdle". In other words, this kind of turbulence undergoes a secondary instability in which the filaments become unstable and roll up into smaller vortices. There are a collection of vortices on the large scale, and these are repeated at the next scale down, and so forth, demonstrating that this type of turbulence has a fractal nature.

Additional Resources

Following are links to a selection of high performance computing and related resources and facilities, most of which reside outside of NYU.

  • NYU HPC Resources for Students
    High performance computing and visualization workstations running Windows and Linux are available for student use at the ITS 14 Washington Place Computer Lab. Call 1-212-998-3058 or 1-212-998-3159 for more information.
  • TeraGrid
    An open scientific discovery infrastructure combining leadership class resources at nine partner sites to create an integrated, persistent computational resource.
  • The National Science Foundation
    An independent federal agency created by Congress in 1950 "o promote the progress of science; to advance the national health, prosperity, and welfare; to secure the national defense.
  • Oak Ridge National Laboratory
    The Department of Energy's largest science and energy laboratory.
  • Argonne National Laboratory
    One of the U.S. Department of Energy's largest research centers. It is also the nation's first national laboratory, chartered in 1946.
  • TOP500 Supercomputer Sites
    A project started in 1993 to provide a reliable basis for tracking and detecting trends in high-performance computing.

Page last reviewed: October 10, 2007