Teraport
Robert Gardner, PI
Goals: The purpose of the Teraport project is to
provide computing and network infrastructure for
a university-based, multi-disciplinary and Gridenabled
data analysis platform with superior
network connectivity to both domestic and
international research networks. The facility has
been deployed and managed as both a local CI
computing resource and as part of larger Grid
and information infrastructures, with specific
17
focus on interoperability with the TeraGrid, Open
Science Grid (OSG), and LHC Computing Grid
(LCG) fabrics.
Significance: The Teraport project was a principal driver to establish the multi-gigabit optical path from the CI to the Starlight facility in Chicago. The facility was first deployed in 2004 and continues to provide major computing resources to local CI users and those accessing the resource via its Grid interface.
Accomplishments: The cluster, which was funded by the NSF and through IBM and University of Chicago contributions, consists of 122 compute nodes with dual 2.2 GHz 64-bit AMD/Opteron processors (244 total), 15 infrastructure/edge service nodes, and 11 TB of fiber channel RAID, all connected with Gigabit Ethernet. The software environment, SuSE Linux Enterprise Server 9 (SLES 9) with the high performance Global Parallel File System (GPFS), was chosen for cost and proven benchmarks of the new Opteron platform and other IBM software components. The Teraport project collaborates with the Biological Sciences Division’s Biomedical Informatics Core, lead by Xiaoming Wang. The Core provides informatics support and integration capacity to improve the effectiveness and efficiency of scientific research by bridging computational expertise to biomedicine. The Teraport project has provided a platform for their two-tier J2EE architecture consisting of front-end web servers and backed Oracle databases hosting biomedical data. Other science applications on Teraport include: Genomic Skimming (Josef Jurek of the Daphne Preuss Laboratory of the Department of Molecular Genetics and Cell Biology); Construction of modular protein-centric databases for biological research purposes (Karl Jablonowski, Ben May Institute for Cancer Research, University of Chicago); Brain research with functional MRI (Uri Hasson, Departments of Neurology and Psychology, Brain Research Imaging Center, University of Chicago); Genome annotation (Robert Olson, ANL/UC); and High Energy Physics (Robert Gardner, UC).
