Integrating High Performance Computing in Research and Education for Simulation, Visualization and RealTime Prediction

Project Details

Project Lead
Anthony Chronopoulos 
Project Manager
Anthony Chronopoulos 
Project Members
Naveen Jayakumar, Yiming Han  
Supporting Experts
Thilina Gunarathne  
Institution
Unvirsity of Texas San Antonio, Department of Computer Sicence  
Discipline
Computer Science (401) 

Abstract

The University of Texas at San Antonio (UTSA) will create the Center on Simulation, Visualization and Real-Time Prediction (SiViRT Center) to integrate the simulation-based research activities in the Colleges of Engineering and Sciences at UTSA, provide computational infrastructure and intellectual leadership for developing outstanding interdisciplinary programs in computationally related research and education, and develop collaborative relations within UTSA, the University of Texas Health Science Center (UTHSCSA) and other regional, national, and international institutions. The SiViRT Center will also attract minority undergraduate and graduate students to engineering and science careers, enhance diversity in UTSA, improve the retention of under-represented minority students and prepare the next generation of scientists and engineers to meet the challenges of solving scientific and engineering problems using reliable, high-performance computational systems and simulations.\n\n\n\nThe SiViRT Center will include three interacting and interdependent research teams in the areas of imaging, real-time prediction and uncertainty quantification. The imaging team will be responsible for the creation of new theory as well as efficient methods and procedures for imaging processes, high fidelity filtering, and image fusion to be used in real-time simulations and control processes. The real-time prediction team will establish a general framework for real-time control and prediction with applications in cancer treatment modeling, surgical control, neuro-biological response simulation and intelligent unmanned vehicles. The uncertainty quantification team will create a general framework for uncertainty quantification in diverse engineering systems and applications of this framework will be sought in engineering fields such as: mechanics of bone fracture; heat transfer with nanofluids; and fate and transport of nanoparticles.\n\n\n\nThe three teams are complementary and support the vision of the SiViRT Center: to be in the forefront of research and to significantly enhance the integration of research and education at UTSA. The multidisciplinary effort on fundamental scientific and engineering research will result in noteworthy technological developments and significant creation of knowledge by bringing together experts from the disciplines of Engineering, Science, Statistics, Biology and Medicine. The Center's educational programs will become the focus of attraction for high school students who will follow science and engineering careers; will provide encouragement and incentives for undergraduate students to pursue research; and will attract, encourage, support, advise, mentor and assist in the placement of doctoral students, who will become the next generation of researchers and university educators.

Intellectual Merit

Intellectual Merit: The proposed project encompasses multiple engineering and scientific areas that are increasingly needed to solve specific problems in the biological sciences. This project will result in a library of methods and algorithms specifically tailored for 3D biomedical imaging. Many ideas from the proposed methods targeting biomedical images are generic in nature and can be applied outside of bio-medical imaging.

Broader Impacts

The impact of the project will be experienced at three levels: students, University and community. One of the major impacts of this project is the broadening of the participation of underrepresented groups in the proposed multidisciplinary research involving engineering and science faculty. The project will strengthen the links between different disciplines in UTSA with the potential of new discoveries. The affiliated faculty will interact with students from engineering and biology, thus sharpening their communication and research skills. Students will be exposed to a research culture in the other disciplines which will help to widen the scope of their interests and

Scale of Use

I need to run comparisons on large systems. However, the time is not large because we develop and \n\ntest scheduling algorithms and the tests last a few minutes (e.g. < 5 mins on e.g. 32 cores).

Results

No results have been obtained.
We struggled for some time to get learn.
Then, we succeeded to build a virtual cluster using nimbus on hotel cluster but we got stuck afterwards because we could not run our (C++, MPI) program because we could not write the job-script correctly. Just to mention that
we have correct job-scripts for running on TACC (ranger). Somehow we could
write the job-script for the virtual cluster on hotel cloud cluster.
We tried to obtain help from the futuregrid-help but their advice did not help us enough
to resolve this.