The San Diego Supercomputer Center (SDSC) at the University of California San Diego has announced the latest round of computer allocations under the Triton Research Opportunities (TRO) program associated with the center's Triton Resource, a data-intensive computing system primarily designed to support UC San Diego and UC researchers.
"The TRO program provides a mechanism for UCSD researchers to access both SDSC's unique computational resources and technical expertise," said SDSC Director Michael Norman. "I am very pleased with the quality and diversity of the interdisciplinary projects so far."
"These latest awards, along with the previous round last March, underscore the intensified levels of collaboration across our campus, as well as the fact that data-intensive supercomputers such as the Triton Resource are becoming a necessary and integral part of scientific advancement and discovery," said Chaitan Baru, SDSC Distinguished Scientist and Chair of the Project Advisory Committee (PAC), comprised of 10 UC San Diego faculty members and researchers who judged the submissions.
Brief summaries of the projects selected from the November 2010 round of submissions, along with researchers from several UC San Diego departments and centers, include:
- Development and Application of Parallel QM/MM Methods for Molecular Dynamics Simulations of Heterogeneous Chemical Reactions on Atmospheric Aerosols
Andreas Goetz and Ross Walker (SDSC), Francesco Paesani (Department of Chemistry and Biochemistry)
The development of high-performance software for simulations that for the first time, would characterize the physicochemical mechanisms associated with heterogeneous reactions on atmospheric particles (aerosols). These reactions play a critical role in influencing the chemical composition of the atmosphere, as well as in altering the Earth's radiative balance, which in turn has major implications for climate, ecosystems, and public health.
- Large Scale Next Generation Sequencing for Cancer Research at SDSC
Sriram Krishnan (SDSC), Karen Messer and Lei Bao (Moores UCSD Cancer Center)
The development and analysis of a large-scale, next-generation sequencing pipeline for cancer research by leveraging the computational infrastructure and knowledge of parallel/distributed algorithms to efficiently maintain and process these large datasets. This project has the potential to lead to significant future research collaborations between SDSC and the Moores Cancer Center.
- Parallel Optimization and Validation of Neuronal Simulations
Kenneth Yoshimoto and Subhashini Sivagnanam (SDSC), Maryann Martone and Vadim Astakhov (Department of Neurosciences)
To develop an effective system for performing large-scale, parallel neuronal computer simulations, including creating general rules for the efficient decomposition of serial neuronal models into parallel ones, and analysis of simulation validity resulting in rules for verifying the consistency of simulations performed under varying conditions.
The latest awards follow a diverse selection of research projects selected from the March 2010 TRO submissions. They include:
- Real‐time Solar Forecasting from Satellite Imagery
DJ Choi (SDSC), Jan Kleissl (Department of Mechanical & Aerospace Engineering) To develop a land-sky-satellite data assimilation system that for the first time would combine these three key elements into a single forecasting system for solar energy, a universally deployable and sustainable resource.
- HI-SEAS: A HIgh-performance Platform for SEmantic Annotation and Search
Amarnath Gupta (SDSC), Maryann Martone (Department of Neuroscience), Shankar Subramaniam (Department of Bioengineering and SDSC), Philip Bourne (Department of Pharmacology and SDSC), Alin Deutsch (Department of Computer Science and Engineering)
This project addresses two related problem areas of scientific knowledge management: semantic annotation and semantic search. The goal is to develop a novel, high-performance semantic annotation and semantic search facility that effectively utilizes the Triton Resource's Data Analysis Facility. Performing semantic annotation and search for scientific information is central to the mission of treating science as a public resource, where information can be effectively linked together to form a knowledge network that promotes information discovery.
- Wave Climatology Responses to Variation in Wind Forcing Using WaveWatchIII
Peter Bromirski and Reinhard Flick (Scripps Institution of Oceanography), John Helly (SDSC, SIO)
To extend previous research using the Triton Resource in the development, analysis and storage of long-term, data-intensive model runs in global wave climatologies, which are used for a variety of research and engineering purposes. The project includes further development on the WaveWatch III code.
- Computational Framework for Cardiovascular Surgical Procedures
Alison Marsden (Department of Mechanical and Aerospace Engineering), Yuri Bazilevs (Department of Structural Engineering), Amitava Majumdar and Mahidhar Tatineni (SDSC)
To enhance current high-performance computing capabilities for cardiovascular modeling, using Triton Resources' HPC capabilities to perform accurate, high-fidelity computational simulations of blood flow, including the vessel deformation of arteries. Such simulations can provide a significant benefit in surgical planning.
- Exploring Page Rank for Data
Dave Nadeau (SDSC), Phil Bourne (Skaggs School of Pharmacy and Pharmaceutical Sciences, SDSC)
To explore data rankingalgorithms that combine multiple data attributes, cross references between data, usage statistics from archives and portals, data citations in the literature, implicit data groupings, and weighting statistics calculated on the data collection as a whole to develop a notion of data page rank. The project is in response to ever-larger online data collections, along with growing lists of search results from user queries, making it difficult for users to find the most authoritative, comprehensive, detailed, popular, widely-referenced, or otherwise "best" data in the collection.
- Development and Application of Structure-based Potentials for Parallel Simulation of a Key Transcriptional Module
Ross Walker (SDSC), Peter Wolynes and Elizabeth Wolynes (Department of Chemistry and Biochemistry)
This collaboration between experts in theoretical biophysics and high-performance computing will use computational models based on the energy landscape theory of folding to simulate the folding/binding reaction of NF-_B transcription factor and inhibitor I_B_, and the I_B_-induced NF-_B•DNA dissociation, called stripping. These interactions are key in the control of many processes of medical importance, such as response to infection, cell death, and cancer.
- A Portal and Database Framework for Large-Scale Computing Collaboration in Earthquake Engineering
Choonhan Youn (SDSC) and Ahmed Elgamal (Department of Structural Engineering)
To develop a robust, extensible searchable shared-usage portal and database for integrating numerical response from thousands of large-scale earthquake engineering computational simulations. Such large-scale simulations will permit much needed probabilistic and sensitivity analyses, greatly facilitating the execution of cutting-edge performance-based earthquake engineering investigations.
The Triton Research Opportunities program provides seed funding for developing collaborative, extramural proposals between SDSC and UC San Diego researchers. The program consists of a proposal competition for 10 percent of Triton Resource cycles and funds equivalent to $200,000 per year in the form of SDSC full-time employees working on the collaborative project. Full details of the TRO program can be found at http://cired.sdsc.edu/tro/troinfo.php.
Deployed in late 2009, SDSC's Triton Resource is a 2,000-processor computing cluster that features some of the most extensive data analysis power available commercially or at any research institution in the country because of its unique large-memory nodes. The system is also available to researchers throughout the larger academic community, as well as private industry and government-funded organizations.
The Triton Affiliates and Partners Program (TAPP) offers various mechanisms for accessing the Triton Resource, which is also available to other universities, research institutes, government agencies, and industry on a space–available basis. Research inquiries may be directed to Ron Hawkins, Triton Affiliates Program Manager, at (858) 534-5045 or rhawkins@sdsc.edu.