What IU information technology systems are connected to the TeraGrid?

The TeraGrid is an open scientific discovery infrastructure combining leadership-class resources at 11 partner sites to create an integrated, persistent computational resource. TeraGrid resources include more than a petaflop of computing capability and more than 30 petabytes of online and archival data storage, with rapid access and retrieval over high-performance networks. Researchers can also access more than 100 discipline-specific databases. This combination of resources makes the TeraGrid the world's largest, most comprehensive distributed cyberinfrastructure for open scientific research.

As a TeraGrid Resource Provider (RP), Indiana University makes these advanced supercomputing, storage, and visualization resources available to the scientific research community:

• Big Red: IU's 20.4 teraflop 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 uses WRF, NAMD, and MILC (IU is 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?

• Quarry: IU's Quarry is an IBM HS21 Bladeserver cluster.

  On the TeraGrid, the Quarry Gateway Web Services Hosting resource (login.quarry.iu.teragrid.org) at Indiana University consists of multiple Dell AMD systems geographically distributed for failover. Each system has at least eight cores and 32GB of memory. Persistent storage is available using IU NFS home directories (10GB default quota) or the 335TB Data Capacitor WAN (Lustre) file system. The system utilizes OpenVZ to provide virtual hosting of RPM-based Linux distributions. The host operating system is Red Hat Enterprise Linux.

  This service is used solely for hosting Science Gateway and Web Service allocations, and is restricted to members of approved projects that have a web service component.

• 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 disaster resilience. For more about 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, email AVL staff.

• Centralized Life Sciences Data service (CLSD): CLSD uses IBM's DB2 database product to federate data from various data sources: 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), as well as selected portions of the public TRANSFAC 2005 data, the Gene Ontology, Uniprot, and the Disease Gene Network (including a protein-protein interaction network constructed in 2005). In addition, it is possible to execute BLAST jobs using data from these databases.

  CLSD makes it possible to write SQL queries that pull data from any of these databases at once, vastly simplifying many data aggregation problems.

  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.

Comments/Questions/Corrections