You must log in to edit wiki content.

Welcome!

This site provides a guide to the capabilities, features and use of the High Performance Computing (HPC) systems supported by UD's Information Technologies' research support staff. You will find details of IT-supported hardware and software environments, instructions and tutorials on their use, and their organizational and administrative structure.

The site also provides links to other high-performance computational resources available to UD researchers, including systems developed and maintained by departments and colleges, as well as IT support for the use national resources such as XSEDE (formerly TeraGrid) and commercial cloud services.

Writing a successful XSEDE allocation proposal (Webinar followed by live discussion) (2013-05-23, anita)

The Extreme Science and Engineering Discovery Environment (XSEDE, formerly called TeraGrid) is a single virtual system that scientists can use to interactively share computing resources, data and expertise. During this Webinar, XSEDE staff will introduce users to the process of writing a successful XSEDE allocation proposal. An allocation is required …

VSCSE summer workshops (2013-05-15, anita)

This July, UD joins a national group of geographically dispersed, prominent universities as host sites for workshops organized by the Virtual School of Computational Science and Engineering (VSCSE). The multi-day virtual workshops will be held in UD videoconferencing studios, led by faculty and researchers from UC San Diego’s Supercomputer …

Systems

UD IT employs a community cluster architectural strategy for building HPC systems: periodically, IT staff engage University faculty in the specification and purchase of a new cluster. The purchase is financed both by IT (cluster infrastructure, storage, and networking) and by faculty buy-in of compute nodes in the cluster.

The following table lists all clusters currently in service. Click on the cluster name to view more detailed information for that cluster.

Inception Date Cluster Name Nodes Processor
Jan 2012Mills20112-core 2.4 GHz AMD 6234 (“Interlagos”)

Documentation

Instructional documentation is stored in general-purpose, cluster-specific, and software-specific areas.

  • Instructions on using the bash shell or on submitting Grid Engine job scripts are documents that pertain to multiple HPC systems. They are found in the general-purpose area.
  • The software-specific area contains local documentation that augments vendor-supplied documentation for application software.
  • The technical documentation area contains information authored by IT-NSS and IT-CSS as the HPC systems are built. This information is published to aid IT staff with future cluster deployments and to aid others who are building HPC systems.

Training

Training materials such as tutorials and workshops are designed to provide our users with instructions on topics pertaining to HPC cluster use in general and IT-supported HPC clusters.

  • A complete list of all training materials by topic can be found in the training area.
  • Specific cluster training materials such as an overview and getting started using the cluster, Mills, can be found in the cluster-specific areas.
  • Training materials specific to using software applications on IT-supported HPC clusters can be found in the software-specific area.
start.txt · Last modified: 2013-03-19 12:26 by anita