TACC Resources

TACC provides comprehensive advanced computing resources, including:

High performance computing (HPC) systems of a variety of architectures to enable larger simulations analyses and faster computation times than are possible using computers available to individual researchers, academic departments, and research centers and institutes;
Advanced scientific visualization (SciVis) resources, including computing systems with high performance graphics hardware, large displays, and immersive environments, and high-end post-production facilities to enable large data analysis and promote knowledge discovery;
Massive data storage/archival systems to store the vast quantities of data that result from performing simulations on HPC systems and developing visualizations of large data sets;
Networking to allow both local- and wide-area access to TACC’s resources; and
Software and tools to assist scientists and technical practitioners in using advanced compute and data hardware resources, and to utilize remote visualization resources.

TACC HPC systems include the following:

Ranger, a Sun Constellation Linux Cluster, is configured with 3,936 16-way SMP compute-nodes (blades), 123TB of total memory and 1.73PB of global disk space. The system components are connected via a full-CLOS InfiniBand interconnect. Eighty-two compute racks house the quad-socket compute infrastructure, with additional racks housing login, I/O, and general management hardware. Compute nodes are provisioned using local storage.
Lonestar, a Dell Dual-Core Linux Cluster contains 5,840 cores (2.66GHz) within 1460 Dell PowerEdge 1955 compute blades, 16 Dell PowerEdge 1850 compute-I/O server-nodes, and two Dell PowerEdge 2950 login/management nodes. Storage includes a globally accessible 70TB Lustre parallel work file system and an aggregate of 106TB of local storage. All compute blades are connected via an InfiniBand network fabric.
Stampede, a Dell Quad-core Linux serial job cluster, consists of 218 compute nodes, two login nodes, and a dedicated file server attached to one of the compute nodes. Each compute node has two quad core Intel Clovertown processors, 8 GB of memory, and 600 GB of local disk space (of which 520 GB is available to the user) and 536 GB of shared disk. The dedicated file server provides 3.7 TB of storage to certain users and is mounted on all of the compute nodes.
Champion, an IBM Power5 System, contains 12 IBM P5 575 server nodes, each with 8–1.9 GHz Power5 processors, and 16GB of shared memory. Server nodes have access to a 7.2TB GPFS file system, and are connected via an IBM Federation switch.

TACC operates an immersive visualization laboratory on the UT Main Campus:

The ACES Visualization Laboratory ("Vislab"), located in ACES 2.404a, is currently under construction. The Vislab is expected to be completed and open to all UT faculty, students and staff by mid-September 2008. The lab will include updated systems and top-of-the-line equipment, such as Stallion, a Dell visualization cluster with 24 nodes and a 9x5 190M pixel tiled display, and a Sony SRX-S105 9M pixel flat projection system driven by a high-end Dell workstation. Click on the link above for more information and a floorplan of the new Vislab.

Data Storage

TACC operates a data storage system comprised of an SGI Origin 2000 running the Data Migration Facility (DMF) software and supported by two StorageTek PowderHorn tape silos with a combined storage capacity of over 2.2 petabytes of user data.

Networking

TACC, the University of Texas at Austin’s ITS Telecommunication and Networking Services, and the University of Texas System’s Office of Telecommunication Services networking infrastructure provides both local- and wide-area communications with TACC resources. TACC also manages a connection to the NSF TeraGrid high-speed backbone.

Software Tools

TACC also creates and modifies software tools to help researchers be more productive when using HPC systems. These tools were developed in response to common user questions, and are intended to aid in cross-platform compatibility and allow researchers to utilize HPC systems in situations where they otherwise might not be able. TACC will continue to enhance existing tools and develop new ones based on usage patterns and user input.