Changes in the landscape of high performance computing reflect constant new developments in hardware and technologies that give HPC machines their computational power. But the programming tools, applications, and algorithms that form the backbone of the ever growing need for greater performance are equally as important. Such myriad hardware/software configurations presents unique challenges that require testing and development of applications that are often quite unique to the platform on which they reside. For this reason, it is imperative that we have access to a very wide range of computing resources in order to conduct our cutting-edge research. On this front, we have multiple, heterogeneous systems in-house. But we also have access to multiple architectures around the country due in large part to our many collaborators and partners. Locally, we maintain systems ranging from individual desktops to large, networked clusters. Below is a summary of the many computing resources used by ICL.

The following are the local systems that we use on a daily basis to test our work:

• 64 node (128 cores) Intel EM64T cluster connected with Myrinet 2000
• 62 node (124 cores) AMD Opteron cluster connected with Myrinet 2000
• 24 node (96 cores) AMD Opteron cluster connected with Silverstorm and Mellanox Infiniband
• 8 node (16 cores) Intel Core2 Duo cluster with GigE
• 8 node (128 cores) AMD Opteron cluster connected with Myrinet 10G
• 5 node (10 cores) Intel Itanium cluster
• SiCortex (72 cores) SC072-PDS

In addition to these resources, we have access to several server class machines and several HPC clusters within the EECS department. These clusters consist of multiple architectures and comprise over 100 machines with various architectures. All of our clusters are arranged in the classic Beowulf configuration in which machines are connected by low latency, high-speed network switches.

Also, exclusive access to many remote resources, some that are regularly found on the Top500 list of the world’s fastest supercomputers, helps keep us at the forefront of enabling technology research. The recent modernization of the DOE’s Center for Computational Sciences, just 30 minutes away at the Oak Ridge National Laboratory (ORNL), has enabled us to leverage our ORNL collaborations to take advantage of what is becoming the world’s fastest scientific computing facility. In addition to an existing Cray XT5 (Jaguar), UT’s National Center for Computational Sciences (NCCS) at ORNL houses UT's Cray XT4 (Kraken), soon to be upgraded to a 10,000+ compute socket XT5. In November 2008, the Kraken system was the 2nd most powerful academic supercomputer in the world and the Jauguar system was the 2nd overall fastest system in the world according to the most recent list of the Top500 supercomputers. The following are some of the remote systems and architectures that we utilize:

• Cray X1E, XT3, XT4, and XT5
• HP XC system
• IBM Power 5, 5+, 6, Cluster 1600, BlueGene/L, and the Cell
• Several large (512+ proc) Linux Clusters
• SGI Altix
• SiCortex