What IU information technology systems are connected to the TeraGrid?

The TeraGrid is the flagship initiative of the US National Science Foundation (NSF) to create a national cyberinfrastructure to accelerate scientific discovery. The TeraGrid focuses on capability science (i.e., things possible only with the biggest supercomputing and storage resources available in the world). The NSF funds Indiana University to be a Resource Provider as part of the TeraGrid, and IU's most advanced supercomputing, storage, and visualization resources are available via the TeraGrid, as follows:

• Big Red: IU's 20.4 teraflops Big Red cluster includes a total of 1,024 dual-core PowerPC processors (two processors per JS21 blade). Big Red is one of the largest supercomputers in North America, and is presently the second largest of the supercomputers accessible via the TeraGrid.

  Big Red is particularly targeted toward the national research community that use WRF, NAMD, and MILC (IU is in the process of developing a version of MILC optimized for the Power Instruction Set). Big Red is an extremely powerful system in its own right. Because it makes use of the Power Instruction Set, you can use Big Red in developing applications that might later be run on Blue Gene systems (which run the same instruction set, although Blue Gene nodes have much less RAM than Big Red). For more information, see At IU, what is Big Red?

• Data Capacitor: The Data Capacitor is a high-speed, high-bandwidth storage system for research computing that serves all IU campuses and TeraGrid users. At peak performance, the Data Capacitor has a 14.5GB per second aggregate transfer rate.

  The Data Capacitor provides high read/write speeds for user data and support for very large files. Using a wide area file system, the Data Capacitor permits users to access remote data as if the file system were mounted locally, allowing you to share large amounts of data with researchers at multiple remote sites.

• HPSS: IU's archival storage system using the High Performance Storage System (HPSS) software has a current capacity of approximately two petabytes. Requests for up to 4.9TB of storage space are granted routinely. See How do I apply for a new TeraGrid allocation? For information about requesting more than 4.9TB, see About applying for a Research allocation on the TeraGrid.

  HPSS is an advanced storage system that supports writing data to tape with encryption. IU's HPSS installation is unique in that researchers can store data to two different storage facilities simultaneously. These facilities are located in Indianapolis and Bloomington, separated by 50 miles, and this provides a good bit of disaster resilience. For more information on using IU's archival storage system, see How do I access IU's MDSS/HPSS from my TeraGrid account?

• BARCO Virtual Reality Theater: The Computer Automatic Virtual Environment (CAVE) was an immersive virtual reality technology consisting of an 8'x8'x8' structure with high-resolution, stereoscopic images projected onto three walls and the floor, a tracking system for the user's head and hand, and an SGI graphics supercomputer (Onyx2). Until it was retired in 2006, the CAVE was installed in Lindley Hall at IU Bloomington.

  A similar but superior immersive technology, the BARCO MOVE Lite Virtual Reality (VR) Theater includes three 8'x8' movable walls. The walls can be used as a walk-in room like a CAVE, or can be rearranged in other configurations. IU's BARCO VR Theater has a pixel resolution of 1600x1200 pixels, which is the greatest resolution of any immersive room-scale visualization system available in the US. The BARCO VR Theater can be powered by an SGI Irix-based system or a Linux cluster.

  Any user with a TeraGrid account and an existing application that uses the CAVE Library (or other 3D software supported by the Barco VR Theater) is welcome to visit IU to use the BARCO VR Theater. This might be particularly useful for people engaged in the development and use of distributed visualization and telecollaboration applications, or computational steering applications. For more information, send email to  avl@indiana.edu .

• Centralized Life Sciences Data service (CLSD): CLSD uses IBM's Information Integrator product to federate data from NCBI: BIND (Pathways, Gene interactions), ENZYME (Enzyme nomenclature), ePCR (ePCR results of UniSTS vs Homo sapiens), SGD (Saccharomyces Genome Database), UniGene, Nucleotide (Nucleotide sequences), PubMed (Journal abstracts), dbSNP, KEGG resources, KEGG data sources, LIGAND (Pathways, Reactions, and Compounds), and PATHWAY (Pathway map coordinates). In addition, it is possible to execute BLAST and HMMR jobs using data from these databases.

  The real advance in CLSD is that a person can do a query that pulls data from any and all databases at once, joins the results properly, and integrates database queries with BLAST and HMMR results. CLSD became available as a TeraGrid resource in 2007.

This document was developed with support from the National Science Foundation (NSF) under Grant No. 0503697 to the University of Chicago and subcontracted to Indiana University. Additional support was provided by IU through its participation in the TeraGrid, which is supported by the NSF under Grants No. 0833618, SCI451237, SCI535258, and SCI504075. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NSF.