Skip to content

Research & Collaboration
  • Home >
  • Building Dedication

SDSC Building Dedication


10.14.08 - San Diego Supercomputer Center Dedicates New 'Green' Campus Building


Transforming Science and Society through Cyberinfrastructure



Level B2: Earthquake Research

TeraShake Earthquake Simulations
Bernard Minster (SIO), Amit Chourasia (SDSC)
Researchers from SDSC, SIO and the Southern California Earthquake Center (SCEC) have conducted some of the largest and most detailed earthquake simulations to date under the TeraShake project, in which ground motions are modeled as expected from a large earthquake on the southern San Andreas fault. The TeraShake calculations simulate 4 minutes of 0-0.5 Hz ground motion in 180,000 km2 area of southern California, for an M 7.7 earthquake along the 199 km section of the San Andreas fault between Cajon Creek north of Los Angeles, and Bombay Beach on the shore of the Salton Sea. The simulations required up to 19,000 CPU hours on 240 processors of the 10 teraflops IBM Power4+ DataStar supercomputer at SDSC. Animations of the simulated wave propagation and synthetic seismograms from TeraShake are available at and

Stressing Buildings for Earthquake Resistance
José Restrepo (JSoE), Amit Chourasia (SDSC) .
To analyze the impacts of an earthquake on a building, engineers studied the building’s structural performance for a slice of a seven-story reinforced concrete residential building, measuring 65 feet high and weighing 275 tons, constructed on a shake table. The testing frequency range was 0-20 Hz. An extensive network of sensors with more than 600 channels was used to measure the dynamic response of the building and its surroundings. These tests yielded a wide range of structural data, including temporal recordings of stress, strain, displacement, accelerations, etc. and

Level B1: Medicine and Life Sciences

The 3D Structure of the Immunoglobulin Heavy-Chain Locus: Implications for Long-Range Genomic Interactions
Suchit Jhunjhunwala (Biology), Cornelis Kees Murre (Biology), Steve Cutchin (SDSC)
The 3D structure of the immunoglobulin heavy-chain (Igh) gene, which codes for antibodies, was determined using 3D-fluorescence microscopy and trilateration. The chromosomal region occupied by the Igh locus spans 3 million base pairs. To probe the topography of the Igh locus, spatial distance distributions were determined between 12 genomic markers that span the entire locus. Comparison of the distance distributions to computer simulations of alternative chromatin arrangements predicted that the Igh locus is organized into compartments containing clusters of loops separated by linkers. This research represented for the first time how a genome is organized in the 3D space.

Cryo-electron Microscopy and 3D Image Reconstruction of Viruses
Robert Sinkovits (SDSC)
Cryo-electron microscopy and three-dimensional image reconstruction are the techniques of choice for determining the structures of large biomolecular complexes that resist crystallization. To help reach our goal of solving virus structures from electron micrographs to the highest possible resolutions in the shortest amount of time, we have developed highly optimized parallel applications and integrated them into an easy-to-use automated image reconstruction system. Computational projects that had formerly taken months or even years of effort can now, in the best cases, be completed in a few hours once boxed image data is available. Overview includes a demonstration of the software system, including examples of virus structures that have been solved by the research group since the start of the collaboration between SDSC and the Timothy Baker Laboratory at UC San Diego’s Department of Chemistry and Biochemistry.

Fighting Influenza Viruses Using Supercomputers
Dong Xu (Chemistry and Biochemistry)
Avian flu has provoked considerable concern since humans have little or no immune protection against the virus. Health officials recognize that new drugs are desperately needed, since some strains of the virus already have developed resistance to the current roster of anti-flu remedies. Now, a team of UC San Diego scientists, including J. Andrew McCammon, holder of the Joseph Mayer Chair of Theoretical Chemistry at UC San Diego, along with researchers from the National Biomedical Computation Resource (NBCR), part of the Center for Research on Biological Systems, and the California Institute for Telecommunications and Information Technology (Calit2), have used SDSC supercomputer simulations to isolate more than two dozen promising and novel compounds from which new “designer drugs” might be developed to combat this disease. The study was recently published in the Journal of Medicinal Chemistry.

Moores Cancer Center Digital Imaging Core and SDSC Collaboration Project
James Feramisco (Moores Cancer Center), Steve Cutchin (SDSC)
The Microscopy Shared Resource of the UC San Diego’s Moores Cancer Center and SDSC have established collaborations to facilitate research in cancer biology and other human diseases. The resource serves to enhance the scientific investigations relating to microscopy and cancer biology, and the biology of other diseases conducted by dozens of members of the Moores Cancer Center and the UC San Diego campus as a whole. Researchers have enhanced research in diverse areas such as breast cancer, prostate cancer, brain cancer, colon cancer, angiogenesis, immunology, gene expression, gene therapy, and cardiovascular disease. Conducting 3-D analysis of positional data within cancer cells or tissues usually generate large amounts of data, and SDSC’s scientific visualization tools, computational power, and secure data storage facilities have been an invaluable part of this research.

Using High Performance Computing to Identify Drug Targets for Parkinson’s Disease
Igor Tsigelny (SDSC), Mark Miller (SDSC), Eliezer Masliah (Neurosciences)
The progression of a large number of human degenerative diseases, including Parkinson’s disease, occurs via assembly of a group of unstructured or partially structured proteins into a structure that is toxic to a specific group of cells. This presentation will review recent Molecular Dynamics Simulations developed at SDSC, describe simulations that predict how assembly of the synuclein proteins at the cellular membrane initiates the molecular events that result in Parkinson’s disease progression, and describe how these simulations can be used to identify pharmaceutical targets to prevent disease progression.

Applications of High Performance Cluster Computing in Radiotherapy Treatment Planning
Steve B. Jiang (Department of Radiation Oncology), Amitava Majumdar (SDSC)
Research at Center for Advanced Computing (CART) is focused on two broad areas of radiotherapy physics: Monte Carlo-based radiotherapy treatment planning, and Image-guided Adaptive Radiation Therapy (IG-ART). Monte Carlo methods are used to model mega-voltage electron and photon beams produced by medical linear accelerators using radiotherapy specific Monte Carlo codes. Since the Monte Carlo method is inherently parallel due to the independent nature of particle transport, the use of parallel processing these simulations helps to reduce overall computational time. The parallelization of these codes has enabled us to perform Monte Carlo calculations within an hour to a few hours using the TeraGrid cluster at SDSC. The other area of research is focused on online adaptive re-planning for cervical cancer patients. Organs in the pelvic region are mobile and distensible. Their daily position and shape vary greatly and tumor geometry usually changes significantly in response to radiation and chemotherapy. Given these factors, treatment plans developed before the therapy are likely to deteriorate in quality and efficacy over time. Therefore, an adaptive therapy scheme is necessary for the treatment of cervical cancer. This should result in improved treatment by explicitly taking into account the daily tumor geometric variation of the patient throughout the course of the treatment. Researchers are collaborating with SDSC to develop an on-line daily adaptive protocol and evaluate the benefits of adaptation on tumor coverage and normal tissue sparing.

Level 1: Green Computing, Data-Intensive Computing

SDSC Data Center Overview
Dallas Thornton (SDSC)
The San Diego Supercomputer Center’s new data center will serve as a living laboratory for efficiency initiatives that may be used at other large data centers around the world. The presentation covers a wide range of energy-saving initiatives being implemented in the new data center, as well as upgrades made to SDSC’s existing data center. In taking a leadership role to implement energy-saving practices across the entire campus, UC San Diego was the first university to join The Green Grid, a global consortium of information technology companies dedicated to promoting energy-efficient data centers.

Green Computing : Energy and Temperature Efficient Computing Using Continuous System Telemetry Harness
Allan Snavely (SDSC), Ayse K. Coskun (CSE), Tajana Simunic Rosing (CSE)
One of the challenges for high-performance computing data centers of the size and computing capacity of SDSC is energy efficiency.  This demonstration shows results from adapting novel telemetry software (monitors temperatures, voltages, currents, and power dynamics vs. load throughout the interior of computer servers) with an intelligent load provisioning scheme to achieve “energy-aware scheduling.” This results in more efficient computations, and additionally reduces the cooling burden, thus lowering thermal gradients inside servers, and improving long-term system reliability.  The demonstration uses a 3D Visualization utility that illustrates thermal and energy dynamics inside SDSC servers during normal workload dynamics. SDSC assisted with the installation of Intelligent Power Monitoring (IPM) software donated for this project by Sun Microsystems, and with the archival of IPM’s “Black Box Recorder” files that capture accurate dynamic thermal flux and power utilization metrics throughout the interior of servers.

Measuring and Visualizing the Global Internet Topology
Young Hyun (SDSC)
What does a snapshot of the Internet look like? The Cooperative Association for Internet Data Analysis (CAIDA) demonstrates its Walrus graph visualization tool employing a fisheye-like distortion to simultaneously display both local detail and global context. The data comes from CAIDA’s ongoing Macroscopic Topology Measurements project. The project actively measures connectivity and latency data for a wide cross-section of the global Internet. To accomplish these measurements, we developed Archipelago (Ark), our next generation measurement infrastructure featuring Marinda, a coordination engine for managing teams of globally distributed monitors. The monitors conduct probes to randomly selected destinations throughout the world. The data describes the paths taken by packets traveling between our monitors and the destinations. The resulting data is collected on a central server at SDSC called the IPv4 Routed /24 Topology Dataset.

David Minor (SDSC)
Chronopolis is a digital preservation data grid framework developed by the SDSC, the Library of Congress, the UC San Diego Libraries (UCSDL), and their partners at the National Center for Atmospheric Research (NCAR) in Colorado and the University of Maryland’s Institute for Advanced Computer Studies (UMIACS). A key goal of the Chronopolis project is to provide cross-domain collection sharing for long-term preservation. SDSC helped in these efforts by providing large-scale storage, high-speed networking, Storage Resource Broker (SRB) installation and support, and other technical services. This work has benefited both the campus community, as well as national users.

UCSD Grid Portal
Shava Smallen (SDSC), Stephen Lyon (SDSC)
The UC Grid is a University of California system-wide grid of software and hardware resources. Using UCLA Grid Portal software, each campus links various machines and software from their campus into an easy-to-use campus Grid portal. Each campus grid portal is then linked to a common UC Grid Portal, allowing resources to be more easily shared with the rest of the UC community. SDSC is collaborating with UCLA to add a UCSD Grid Portal to the UC Grid. Currently, the UCSD Grid Portal contains a 7-node test cluster and plans to add SDSC production resources, Blue Gene and OnDemand, within the next few months.

Level 2: Technology for Education

Phil Bourne (Pharmacology), Apryl Bailey (SDSC)
SciVee invites scientists to make their research known by combining their published scientific article with a corresponding video into an online presentation called a SciVee pubcast. The community can then freely view the presentation and engage in virtual discussions with you and other SciVee members about the research directly through commentary, community discussions, and blogging features. SciVee also encourages everyone interested in science to create a free membership and join existing scientific communities, or create their own new community around specific scientific interests where discussions and events can be organized, shared, and documented.

Ubiquitous Presenter
Beth Simon (CSE), Roshni Malani (CSE)
Ubiquitous Presenter supports faculty by bringing the best of the chalkboard to the digital classroom experience. Using a Tablet PC, faculty can ink over diagrams and other prepared material and have all notes stored online for reviewing. Additionally, UP provides support for real-time interactive classrooms (polling, text submissions, and ink submission), allowing hundreds of students to send in answers for immediate review (and projection) in class.

Information Technology – Engineering and Environmental Education (IT-E3) Tools
Jeanne Ferrante (JSoE), Diane Baxter (SDSC)
UC San Diego’s Jacobs School of Engineering and SDSC are launching an environmental education project designed to entice middle school girls to pursue science and engineering. The team is teaching San Diego County teachers and students to monitor the air quality, solar radiation, and other environmental factors surrounding their own schools. The same environmental research concepts and techniques factor into a multi-player online science challenge game designed specifically for 12-15 year-old girls. IT-E3 Tools is funded through a three-year, $1.2 million National Science Foundation grant as part of their Information Technology Experiences for Students and Teachers (ITEST) award program. This program was created in response to the shortage of information technology workers in the U.S. and is supported with supported by income from ‘H-1B’ visas awarded to professionals from other countries recruited to fill specialized jobs in the U.S.

Town and Country Village Learning Center
Michael Cole (SOM), Diane Baxter (SDSC)
UC San Diego students and faculty, along with SDSC staff, have been working with the San Diego Community Housing Corporation to create an after-school enrichment program at a HUD housing development learning center. The program integrates technology into a wide variety of activities to help participants learn to use technology to enrich their learning, express their creativity, and enable connections with youth from around the world. Each quarter, the program involves approximately 12 to 16 UC San Diego undergraduates, 5 graduate students, and 50 schoolchildren who live in the development, fairly evenly distributed among elementary, middle, and high school. Projects within the program use media as a doorway to the world, teaching them to collaborate with one another and learn about other places in the world. The UC San Diego Medical School and CalIT2 are joining the program this fall, expanding it to investigate ways to use technology tools to help create healthier lifestyles and safer neighborhoods for the residents of this San Diego community.

Assignment Antarctica: Serious Gaming for Environmental Education
Jeff Sale (SDSC), Jeanne Ferrante (JSoE)
Assignment Antarctica is a 3D multi-player game that lets students learn about scientific research in the Antarctic through role-playing as scientists who explore the environment and collect various types of data, including ultra-violet radiation, temperature, and water quality parameters. SDSC is collaborating with UCSD’s Jacobs School of Engineering to develop and host the game for teachers and their students throughout San Diego county. Within the environment players can navigate, communicate, explore, measure, interact and manipulate a host of objects and materials as they seek to solve an environmental challenge. Under development are 3D persistent gaming client using a widely available engine, a reusable persistent gaming server, a web application that provides full review of tracked in-game statistics in real time with game activity, and a data toolkit to simplify instrumenting any other game to record player activity statistics for a wide array of gaming environments. These tools will be freely available from the project web site.

Level 3: Climate Change, Our Origins

CARTA Project
Chaitan Baru (SDSC), Margaret Schoeninger (Anthropology), John Moreland (SDSC)
A multidisciplinary group of researchers at the UC San Diego, along with colleagues at the Salk Institute for Biological Studies in La Jolla, have established a center to formally explore the origins of humanity and the many facets of what makes us human.  CARTA (Center for Academic Research and Training in Anthropogeny) is based on an effort that has been underway for more than 10 years. Goals include understanding more about human and primate genetics and evolution, while advancing relevant knowledge in areas such as language, communication and cognition, human society, and culture. The initiative is currently being converted into a larger and more public research program that will also facilitate graduate and undergraduate education.

High-Resolution Light Detection and Ranging (LiDAR) Topography Data: Visualizations in Google Earth and iView3D
Christopher Crosby (SDSC), Debi Kilb (SIO)
In 2005, the “B4 Project” collected high-resolution Light Detection And Ranging (LiDAR) topography for 1,000 km of the southern San Andreas and San Jacinto Faults in southern California. These data 
provided pre-earthquake topography necessary to determine near-field ground deformation after a future large earthquake (hence the name B4), and support tectonic and paleoseismic research. At the Scripps Institution of Oceanography (SIO), the Satellite Geodesy group and SIO Visualization Center have created 3D visualizations of the B4 project data that can be interactively viewed in Google Earth or the iView3d freeware package. The GEON Project, a cyberinfrastructure for the geosciences project at SDSC, has emphasized the development of tools to distribute and process LiDAR datasets such as B4. In collaboration with the SIO Visualization Center, GEON is incorporating the SIO Google Earth and iView3D data products into their topographic data portal to provide integrated access to data, processing, and visualizations.

Global Climate: Research Overview of Dynamic Downscaling for California and Other Regions
John Helly (SDSC), Masao Kanamitsu (SIO), Kei Yoshimura (SIO)
Currently available meteorological data is only in 200~300 km scale horizontally. In this scale, the central valley is presented as just one grid. To obtain more precise meteorological and climatic phenomenon in finer scales, this project ran a regional model in a 10 km horizontal scale. Those micro-scale meteorological fields are useful for application studies for agriculture, wild fire prediction, wind power generation, water resources management, etc. SDSC storage and computing resources have allowed researchers to create these downscaling projects in a much shorter time -- it takes approximately 140 hours (excluding waiting time) with 256 CPUs and 50GB disk storage for a one-year 10km downscaling over California.

Exploring OptIPortals as Petascale Simulation End-Stations
Michael Norman (SDSC), Rick Wagner (Physics)
The Laboratory for Computational Astrophysics (LCA) and SDSC, working in collaboration with CalIT2, are exploring the potential of OptIPortals to serve as visualization and analysis end-stations for multi-TB data sets arising from petascale simulations carried out on TeraGrid resources, principally the Track 2 systems at the Texas Advanced Computing Center (TACC) and the University of Tennessee/Oak Ridge National Laboratory’s National Institute for Computational Sciences (NICS.) OptIPortals, developed by CalIT2, are scalable display walls tightly integrated with a specially configured Rocks-managed Linux cluster and driven by optical lambdas connected to local and/or remote disk caches. Up until now, OptIPortals have been mainly used for displaying data from instruments, as well as remote collaboration portals. LCA, SDSC, and CalIT2 have begun a 12-month pilot project in which an OptIPortal will be built and integrated into the TeraGrid. As an initial application driver, we will focus on ENZO data, which is being produced at prodigious rates (10’s TB/mo.) on both Ranger and Kraken through the PI’s LRAC allocation.

HiSeasNet: Providing Internet Service to Oceanographic Research Ship at Sea
Jonathan Berger (IGPP)
Have you ever wondered about the large “ears” on SDSC’s roof? They form the hub of HiSeasNet, a joint effort between SDSC and the Scripps Institution of Oceanography to provide continuous Internet connectivity for oceanographic research ships and platforms. While access to the Internet is an integral part of nearly every research lab and office on land, extending this access to oceanographic ships -- our seagoing laboratories -- is a fairly recent innovation. For the ships, HiSeasNet provides transmission of hot data in real-time to shore-side collaborators; basic communications (e-mail, voice, and video teleconferencing) for scientists, engineers, and crew at sea; and tools for real-time educational interactions between shipboard scientists, teachers, and the classroom, as well as informal science and other education and outreach activities.

Hans-Werner Braun (SDSC), Frank Vernon (SIO)
Live demonstrations include: Cabrillo National Monument intertidal area, Palomar Observatory, Santa Margarita Ecological Reserve
The High Performance Wireless Research and Education Network (HPWREN) team will be showcasing several remote science sites via their pilot work on Live Interactive Virtual Explorations (LIVE). The primary objectives of HPWREN LIVE are to explore and understand hard-to-reach science sites, and prepare students for going on field trips to such sites. Specific LIVE activities for this demonstration include the Palomar Observatory, the Santa Margarita Ecological Reserve, and the Cabrillo National Monument intertidal area. HPWREN is a National Science Foundation funded network research project, which also functions as a collaborative cyberinfrastructure on research, education, and first responder activities. It includes creating, demonstrating, and evaluating a non-commercial, prototype, high-performance, wide-area, wireless network in San Diego, Riverside, and Imperial counties. The network includes backbone nodes at the UC San Diego and San Diego State University campuses, and a number of “hard to reach” areas in remote environments.

Using a Mac Laptop and iSeismograph to Record and View Synchronized Acceleration and Video Data
Charles Cowart (SDSC), Lelli Van Den Einde (JSoE)
Based at the San Diego Supercomputer Center, the NEES Cyberinfrastructure Center (NEESit) is a service-focused organization created to deliver information technology tools and infrastructure to enable earthquake engineers to remotely participate in experiments, perform hybrid simulations, organize and share data, and collaborate with colleagues. NEESit developed iSeismograph, an open source Mac software program, which can be used on a Macintosh laptop to record and view real-time acceleration measurement and synchronized video for earthquake engineering research and education. Current Macintosh laptop computers are equipped with iSight cameras and a Sudden Motion Sensor (SMS), a 3-axis accelerometer that detects sudden motion and enables the hard drive to “freeze” and save its contents. With proper calibration, a Mac laptop can become a flexible, cost-efficient, portable education and outreach tool. Highlights of iSeismograph include real-time acceleration data acquisition and video capture, real-time archiving and history replay, and synchronization between sensor data and video images.

back to top