Workshop Program - Center for Computational Sciences

CyberBridges 2013 – Developing the Next Generation of Cyberinfrastructure Faculty for Computational and Data-enabled Science and Engineering


Welcome to the NSF CyberBridges 2013 Workshop!

Overview of the NSF CyberBridges Workshop

The NSF CyberBridges 2013 Workshop has been designed to bring together the community of NSF OCI/ACI CAREER researchers to initiate new collaborations, encourage networking, and to provide feedback to NSF on how to further build the community of cyberinfrastructure researchers. The workshop offers a scientific program in five critical areas of cyberinfrastructure that include: Interdisciplinary Research and Grand Challenges in Cyberinfrastructure, Computational- and Data-enabled Science and Engineering, High Performance Computing, Visualization, and Education.

Over the next two days, leaders from each of these areas will present keynote talks on topics ranging from data enabled molecular modeling and visualization to high performance computing. NSF program directors will also give presentations on programs that involve cyberinfrastructure and will participate on a panel which will offer insight into developing a research program in cyberinfrastructure. Another highlight of the program is the poster session, in which NSF CAREER Awardees will present posters featuring topics such as studies of 3D dynamics in the global magnetosphere, multiscale sensing and simulations for bridge scour, scalable communications, high performance computing and visualization in earthquake modeling, as well as other topics. The workshop also offers several opportunities for networking and discussion of cyberinfrastructure challenges.

We hope that you will find the workshop useful for meeting new colleagues, developing new research connections, and gaining new insights in developing a research and education career in cyberinfrastructure.

Thomas Hacker and Suzanne Shontz Co-Chairs, NSF CyberBridges Workshop


Agenda

(All presentations are to be held in the Main Ballroom.)

Monday, July 15th

07:30-08:30 Breakfast - Lobby 08:30-09:00 Farnam Jahanian, Assistant Director for CISE, National Science

Foundation 09:00-09:30 Alan Blatecky, Division Director, Advanced Cyberinfrastructure,

National Science Foundation 09:30-10:00 Break - Lobby 10:00-10:30 Computational- and Data-enabled Science & Engineering

Omar Ghattas, University of Texas – Austin 10:30-11:30 Discussion 11:30-12:30 Lunch - Lobby 12:30-13:00 High Performance Computing

William Gropp, University of Illinois - Urbana-Champaign 13:00–14:00 Discussion 14:00-14:30 Break - Lobby 14:30-15:30 Program Director Panel 15:30-16:00 Break - Lobby 16:00-17:30 Poster Session 18:00-20:00 Dinner (Junior Ballroom)

Tuesday, July 16th 07:30-08:30 Breakfast - Lobby 08:30-09:00 Education

Steve Gordon, Ohio Supercomputing Center 09:00-10:00 Discussion 10:00-10:30 Break - Lobby 10:30-11:00 Data Enabled Molecular Modeling, Uncertainty Quantification and

Visualization Chandrajit Bajaj, University of Texas - Austin

11:00-12:00 Discussion 12:00-13:00 Lunch - Lobby 13:00-13:30 Grand Challenges in Cyberinfrastructure & Interdisciplinary Research

Brian Athey, University of Michigan 13:30-14:30 Discussion 14:30-15:00 Workshop Summary

15:00 Departure


Speakers/Presentations

Monday, July 15, 2013 8:30-9:00am Emerging Frontiers in Computing and Communication Farnam Jahanian, Assistant Director for CISE, National Science Foundation The mission of NSF’s Directorate for Computer and Information Science and Engineering (CISE) is to uphold U.S. leadership in computing, communications, and information science and engineering. To achieve this, CISE supports investigator-initiated research in computer and information science and engineering, fosters broad interdisciplinary collaboration, helps develop and maintain cutting-edge national cyberinfrastructures for research and education, and contributes to the developments of a computer and information technology workforce with skills essential for success in the increasingly global market. This talk will focus on the technological advances and emerging frontiers that are accelerating the pace of discovery and innovation across all science and engineering disciplines and how they inform NSF’s investments. In particular, it will describe the CISE Directorate’s current initiatives and emerging priorities, including new research opportunities. These efforts provide a foundation for economic competitiveness and will drive new innovations supporting our national priorities, such as sustainability, smart transportation, disaster resilience, education and life-long learning, public safety and national security. 9:00-9:30am Alan Blatecky, Division Director, Advanced Cyberinfrastructure, National Science Foundation 10:00-10:30am - Computational- and Data-enabled Science & Engineering Big Data Meets Big Models: Towards Solution of Large-Scale Bayesian Inverse Problems Omar Ghattas, John A. and Katherine G. Jackson Chair in Computational Geosciences, Professor of Geological Sciences and Mechanical Engineering, and Director of the Center for Computational Geosciences in the Institute for Computational Engineering and Sciences at the University of Texas at Austin One of the greatest challenges in computational and data-enabled science and engineering (CDS&E) today is how to combine complex data with large-scale models to create better predictions. This challenge cuts across every application area within CDS&E, from the geosciences to materials to chemical systems to biological systems to astrophysics to engineered systems in aerospace, transportation, buildings, and biomedicine, and beyond. At the heart of this challenge is an inverse problem: we seek to infer unknown model inputs


(parameters, source terms, initial or boundary conditions, model structure, etc.) from observations of model outputs. The critical need to quantify the uncertainty in the solution of such inverse problems has gained increasing recognition in recent years. This can be carried out in a coherent manner by casting the problem as one in Bayesian inference. Here, uncertain observations and uncertain models are combined with available prior knowledge to yield a probability density as the solution of the inverse problem, thereby providing a systematic means of quantifying uncertainties in the model parameters. This facilitates uncertainty quantification of model predictions when the resulting input uncertainties are propagated to the outputs. Unfortunately, solution of such Bayesian inverse problems for systems governed by large-scale, complex computational models with large-scale, complex data has traditionally been intractable. However, a number of advances over the past decade have brought this goal much closer. First, improvements in scalable forward solvers for many classes of large-scale models have made feasible the repeated evaluation of model outputs for differing inputs. Second, the exponential growth in high performance computing capabilities has multiplied the effects of the advances in solvers. Third, the emergence of MCMC methods that exploit problem structure has radically improved the prospects of sampling probability densities for inverse problems governed by expensive models. And fourth, recent exponential expansions of observational capabilities have produced massive volumes of data from which inference of large computational models can be carried out. 12:30 -1:00pm - High Performance Computing William Gropp, Thomas M. Siebel Chair in Computer Science, Computer Science Department; Director, Parallel Computing Institute; Deputy Director for Research Institute for Advanced Computing Applications and Technologies at University of Illinois – Urbana-Champaign 2:30-3:30pm – Program Director Panel Evelyn M. Goldfield, Program Director, Chemistry Division, National Science Foundation Daniel S. Katz, Program Director, Division of Advanced Cyberinfrastructure, National Science Foundation Peter McCartney, Program Director, Division of Biological Infrastructure, National Science Foundation Thomas Russell, Senior Staff Associate, Office of the Assistant Director, Mathematical and Physical Sciences, National Science Foundation Barry I. Schneider, Program Director, Division of Advanced Cyberinfrastructure, National Science Foundation 4:00-5:30pm – Poster Session (See listing on next few pages)


Tuesday, July 16, 2013 8:30-9:00am – Education Steven I. Gordon, Interim Co-Executive Director, Ohio Supercomputer Center 10:30-11:00am - Data Enabled Molecular Modeling, Uncertainty Quantification and Visualization Chandrajit Bajaj, Director, Center for Computational Visualization, Institute for Computational and Engineering Sciences and Professor of Computer Sciences at University of Texas at Austin Discoveries in bioinformatics promise to revolutionize the treatment and prevention of diseases. With the rapid growth and availability of sequence and structural data for thousands of proteins, nucleic acids, our inability to effectively utilize this is a major hurdle in our obtaining a comprehensive understanding of the function of bio-molecules. In this talk, I shall focus on data enabled modeling to structure elucidate, predict, validate and visualize molecular-molecular structural interactions for use in molecular therapeutics. I shall focus on a core library of sophisticated algorithms for processing multi-modal imaging and sequence data, molecular simulations, and performing the non-convex optimization required for scientific based drug discovery. These algorithms make use of some apriori knowledge about the specimen and a characterization of the uncertainty (stochasticity) in the data which arises from multiple experimental sources (e.g. gene arrays, x-ray diffraction, NMR, and electron microscopy). The choice and models of the target, the choice and search through appropriate drug databases, and the selection of the top leads for druggability all require additional computational analysis and assessment of sequence, structure and function interactions, including effective methods for visualizing the proposed solutions and interactively tuning the drug lead optimization process. Examples of using these methods to rapidly predict new potential therapeutics include those for HIV, Machupo (Bolivian hemorraghic fever virus, also known as the Black Typhus) and other highly infectious diseases like Dengue fever and African Sleeping Sickness. 1:00-1:30pm - Grand Challenges in Cyberinfrastructure and Interdisciplinary Research Brian I. Athey, Collegiate Professor and Inaugural Chair of the Department of Computational Medicine and Bioinformatics at the University of Michigan Medical School


Poster Session:

1. Open software, GPU computing & flipped classroom - Lorena Barba, George Washington University The title expresses three features that are distinctive of my research and educational activity. My CAREER project centers around work with a family of algorithms that fall under the name 'fast multipole methods', or FMM. Several groups are working on FMM, and it has proved to be a hot topic in the last couple of years. But my research group has consistently carried out this work "in the open"—we started with PetFMM (2009), then published GemsFMM (2010) to accompany a book chapter, then 12stepFMM (2011) for a summer school and finally exaFMM (2012), and we are now working on a new code. Always, codes are open-source under the MIT license. Across different student projects, all students work with GPU hardware and we currently are working on two applications using GPUs: computational fluid dynamics (cuIBM), and biomolecular electrostatics (PyGBe). In the educational front, I developed and taught twice a course in computational fluid dynamics under the flipped classroom model (a variant of blended or hybrid learning). My lecture videos for this course on You Tube collected more than 100,000 views in just over a year.

2. TOLKIN-Workflow: A web application for conducting and managing complex research pipelines - Nico Cellinese, University of Florida TOLKIN, the Tree of Life Knowledge and Information Network (www.tolkin.org), is a web-based application that allows remote collaboration among biodiversity scientists. TOLKIN provides informatics support for and facilitates shared-access to a variety of digital data that include voucher specimens, morphology information, nucleic acid samples and sequences, chromosome data, images, etc. The large amount of information stored in TOLKIN creates a demand for data synthesis and analysis. Implementing a series of analytical pipelines or workflows fulfills this need. Unfortunately, pipelines can be hard to assemble by the average biologist and are often executed by command-line based programs. We propose a TOLKIN-Workflow module to assist TOLKIN users in executing and managing complex pipelines. A variety of tasks can be designed, implemented and stored in a workflow library. Task details include input-output requirements, parameter settings, and various other metadata. Users may specify parameters according to the analytical requirements. Similar to stacking individual building blocks, each of these smaller tasks can be reused to build larger workflows. The workflows created in TOLKIN can also be exported in a workflow modeling language compatible markup file for use in alternative scientific workflow software. Additionally, an important feature of TOLKIN-Workflow is the ability to design each task as a RESTful-based service, which can include and utilize distributed computing resources on demand. Our architecture and user-interface provide the flexibility and scalability for a TOLKIN user to execute complex analyses run by


automated pipelines within the same TOLKIN platform where data are stored. This model facilitates the creation, execution, and sharing of complex analyses, including metadata and tracking of data provenance. TOLKIN-Workflow is free and open source under the GPLv3 license. The source code can be downloaded from https://github.com/FLMNH-Informatics/Workflow-RESTful.

3. Discoveries in compressible turbulence and turbulent mixing through Petascale simulations and analysis - Diego A. Donzis, Texas A&M University Compressible turbulent flows, characterized by a wide range of spatial and temporal scales, occur in many natural (supernovae explosions, volcanic eruptions, solar wind) and engineering systems (commercial and supersonic flight and propulsion, reacting flows). Due to their complexity, though, our fundamental understanding remains very limited, especially at high Reynolds numbers, restricting our ability to predict natural phenomena and design better engineering devices. This research advances the knowledge of compressible turbulence only possible through massive simulations enabled by Cyberinfrastructure pushing, in turn, the frontiers of computational fluid dynamics. The simulations, unprecedented in size and detail, require novel software that efficiently uses hundreds of thousands of processors at Petascale levels and set the path for simulations at even larger scales. The emphasis is in extreme parallelism exploring advanced features likely to exist in future architectures including communication libraries, programming models and accelerators. The tremendous details unveiled by the simulations provide a unique opportunity for investigators to understand long-standing issues such as small-scale universality, inertial-range scaling, intermit tency and largely unexplored areas such as mixing and dispersion of contaminants. Results will be made available through a portal using Cyberinfrastructure, which allows the community to obtain flow fields, statistics and codes, and analyze massive datasets remotely. Research and education are integrated at the interface between high-performance computing (HPC) and fluid mechanics. The educational goal is to (i) inspire students, including those from underrepresented groups, to pursue careers in HPC and fluid mechanics, and (ii) educate future scientists and engineers capable of deploying Petascale solutions to important societal and technological problems. Educational and research activities are integrated through curricular innovation, undergraduate research and the portal which is also used to disseminate results to a broader audience.

https://github.com/FLMNH-Informatics/Workflow-RESTful


4. High Performance Computing and Visualization In Earthquake Modeling Research and Education - Benchun Duan, Texas A&M University Megathrust earthquakes along subduction zones may be the most devastating disasters to modern society, as evidenced by the 2011 M9.0 Tohoku (Japan) and 2004 M9.1 Sumatra earthquakes. They not only cause strong shaking in surrounding areas, but also are capable of generating deadly tsunamis that are beyond imagination. In this career project, we will develop dynamic rupture models of the 2011 Tohoku earthquake using our finite element method (FEM) code EQdyna to explore what geological features and processes controlled the rupture behavior of the earthquake. We will also further develop EQdyna to simulate earthquake cycles to investigate the relationship between the 2011 event and other M7~8 earthquakes along the subduction zone, and to explore physical controls on large earthquakes of different sizes along subduction zones worldwide, including the Cascadian subduction zone along the west coast of North American continent. EQdyna has been parallelized using a hybrid MPI/OpenMP approach, and high performance computing on large-scale multi-core supercomputers and visualization of modeling results will be important components in these research activities. We will develop course modules on earthquake generation processes, including inter-seismic slow deformation, nucleation of rupture, dynamic rupture propagation, and transient post-seismic deformation, for undergraduate and graduate earth science curricula, and tailor them for high school earth/space science and physics curricula. Visualization of these processes within the 3D Earth is an efficient way to convey scientific knowledge to students. We will also build physical spring-slider models to illustrate earthquake interaction and complexity.

5. Non-equilibrium quantum dynamics in strongly correlated systems –Adrian Feiguin, Northeastern University We apply the time-dependent density matrix renormalization group (DMRG) technique to the study of quantum dynamics in strongly correlated quantum systems. The method has become highly instrumental in the study of non-equilibrium, and we use it to tackle a number of topical open problems, such as quantum information processing, decoherence in open quantum systems, non-equilibrium transport in mesoscopic systems, and dynamics of cold atomic systems.

6. Petascale DNS of evaporating droplet-laden homogeneous turbulence - Antonino Ferrante, University of Washington

We are developing a numerical methodology and a petascale flow solver for performing direct numerical simulations (DNS) of evaporating droplet-laden homogeneous turbulence with the objective to study the effects of evaporating droplets on turbulence.


7. A Continuum-mechanics Based, Cyber-enabled Approach to Unraveling Process-structure-property Relationships in Organic Electronics - Baskar Ganapathysubramanian, Iowa State University Developing a computational framework to model the physics of morphology evolution during solution based processing of organic solar cells. Explore process-structure-property relationships to enable design of better solar cells.

8. Studies of 3D dynamics in the global magnetosphere using high-performance heterogeneous computing architectures – Kai Germaschewski, University of New Hampshire Magnetic reconnection and secondary instabilities of thin current sheets are of crucial importance to understanding the dynamics of Earth’s global magnetosphere and space weather. We will present OpenGGCM global magnetosphere simulations of 3-d magnetopause reconnection at the Earth's day side, with a particular focus on effects of extended MHD models. We will also show results on the formation of mirror mode structures in the magnetosphere using local kinetic (particle-in-cell) simulations. We will report on progress in porting our codes to heterogeneous computing architectures using automated code generation.

9. Fault Analysis and Reliability Guided Scheduling for Large-scale HPC Systems - Thomas

J. Hacker, Purdue University Parallel applications that use large-scale HPC systems suffer from the poor reliability of those systems arising from the inherent low reliability of commodity hardware components and software. The impact of this problem is that overall system efficiency is limited by the need for reactive fault tolerance techniques such as checkpointing and job restart. Our research is focused on developing new approaches for fault avoidance and fault tolerance for large-scale HPC systems to improve the resilience of parallel applications. The objectives of our work are to: 1) increase the accuracy and speed of fault detection and prediction for large-scale systems; 2) create new approaches to proactively respond to potential faults and recover from faults; 3) investigate architectures for a small-scale cluster fault injection framework and testbed to assess fault prediction, detection, and recovery mechanisms; and 4) create and deliver an education and training program to increase awareness of use of current and new fault-aware practices. In our poster, we describe a brief summary of our efforts, and identify some emerging challenges and areas for future work in developing more resilient large-scale systems and parallel applications.


10. Traveling Fires – Do They Really Matter? – Ann Jeffers, University of Michigan Ann Arbor The research objective of this NSF CAREER proposal is to characterize the influence of traveling fires on the fire resistance of steel-framed buildings using a simulation-based approach. In particular, a high-resolution fire model will be coupled to a 3D structural model, and parametric studies will be conducted to assess the performance of a prototype steel-framed building given variations in the compartment geometry and ventilation, the type and distribution of the fuel, the location of fire origin, the structural detailing, and other key parameters. The coupling of a high-resolution fire model to a structural model requires consideration for the transport of data across disparate temporal and spatial scales, thus necessitating the use of novel heat transfer finite elements, subcycling, and data homogenization algorithms to enable an accurate, efficient, and scalable simulation of the coupled response. Numerical studies of a prototype building under various traveling fire scenarios will provide new insight into the mechanical actions that develop in buildings under non-stationary fire exposures, allowing for improvements in the design process to account for detrimental effects that may be attributed to traveling fires. The education plan includes: (i) the creation of a hands-on museum exhibit regarding building fire safety that will be installed in the Michigan Firehouse Museum, (ii) mentored research experiences for undergraduate and graduate students that specifically target women and underrepresented minorities, (iii) educational research to assess the learning outcomes of international service-learning in engineering education, (iv) the translation of the research findings to technical activity committees that are concerned with the development of performance-based standards for structural fire design, and (v) an international workshop/symposium to bring together experts from the fire sciences and structural engineering disciplines to address complex issues that exist at the fire-structure interface.

11. Developing middleware to support Distributed Dynamic Data-intensive (D3) science on Distributed Cyberinfrastructure (DCI) – Shantenu Jha, Rutgers University This CAREER Award project will develop middleware to support Distributed Dynamic Data-intensive (D3) science on Distributed Cyberinfrastructure (DCI). Existing NSF-funded CI systems, such as the Extreme Science and Engineering Discovery Environment (XSEDE) and the Open Science Grid (OSG), use distributed computing to substantially increase the computational power available to research scientists around the globe; however, such distributed systems face limitations in their ability to handle the large data-volumes being generated by today's scientific instruments and simulations. To address this challenge, this project will develop and deploying extensible abstractions that will facilitate the integration of high-performance computing and large-scale data sets. Building on previous work on pilot-jobs, these new abstractions will implement the analogous concept of "pilot-data" and the linking principle of "affinity."


The result will be a unified conceptual framework for improving the matching of data and computing resources and for facilitating dynamic workflow placement and scheduling. This research has the potential to significantly advance multiple areas of science and engineering, by generating production-grade middleware for accomplishing scalable big-data science on a range of DCI systems.

12. Topology Optimization of Linear and Nonlinear Multiscale Systems – Kapil Khandelwal, University of Notre Dame Design of structural-material multiscale systems that perform optimally under given loading environment continues to be a challenging task. Since fabrication/manufacturing and testing of novel designs is a time consuming process and can be prohibitively expensive, computational topology optimization methods present a viable alternative. Topology optimization is a mathematical process of finding optimal layout of materials in a given domain, with the aim of optimizing desired performance objectives. This poster presents a computational paradigm including: hardware, computational and visualization algorithms used for topology optimization of multiscale systems. The outreach activities associated with the CAREER project are also presented.

13. SMART: Scalable Adaptive Runtime Management Algorithms and Toolkit – Andy Li, University of Florida Emerging general-purpose high-end computing and storage resources provide great opportunities to tackle grand challenge problems in science and engineering. However, most systems do not have adequate support to meet special requirements of a number of ultra-scale dynamic scientific applications due to lack of application-awareness. These computation and data intensive applications are aimed to model and investigate highly dynamic and sometimes drastically changing phenomena in science and engineering, such as interacting black holes, global and regional high-resolution weather forecasting, combustion and detonation simulation, and many others. Furthermore, it is challenging, error-prone, and time-consuming for scientists/engineers to develop their large-scale parallel and distributed scientific applications from scratch. SMART fills the gap and aims to (1) create an integrated framework of scalable adaptive runtime management algorithms, libraries, and toolkit with friendly programming models so that scientists can write sequential programs to achieve automatic parallelism and high performance and throughput; (2) design a suite of application-aware adaptive algorithms to holistically address various issues in computation, communication, data, and energy management in systems with thousands of processors (such as clusters, grids, and clouds); and (3) enable high-impact real-world large-scale scientific applications with additional tools for simulation and visualization.


14. Modeling Made Easy – Laurence Loewe, University of Wisconsin - Madison Mathematical models are frequently presented in a form that makes it difficult for non-mathematically trained persons to understand (let alone build) them. I will present an overview of Evolvix, a new model description language that I am designing to make model building easy. Evolvix aims to maximize readability for humans and minimize the amount of information modelers have to specify in order to describe a model they want to analyze. This language could be useful for a range of activities from teaching to developing cutting edge stochastic and deterministic simulation models and encoding them for storage in the online supplements of journal articles that present the results from such models.

15. Algorithmic and Software Foundations for Large-Scale Graph Analysis – Kamesh Madduri, Penn State University Graph abstractions and graph-theoretic computations play a critical role in the analysis of data from experimental devices, medical data sets, socially-generated data from the web and mobile devices, and scientific simulation data. Motivated by current terascale applications in genomics, proteomics, and social network analysis, the project will undertake the clean-slate design of four algorithmic frameworks that capture broad classes of graph-based computations: traversal-based static graph computations, dynamic graph analytics, subgraph enumeration and pattern search computations, and multiscale graph computations. This research will lead to the design of new memory-efficient graph representations and data structures, the creation of novel linear time parallel algorithmic strategies based on data partitioning, and a deeper understanding of architectural features that impact graph processing efficiency and scalability. The poster will elaborate on recent research results related to the subgraph enumeration and counting framework on multicore platforms. The proposed educational activities attempt to foster an environment of CDS&E within Penn State. New graduate classes on parallel graph analysis, parallel algorithms for computational biology, and high-performance social data mining, will facilitate student involvement in current research activities of this project.

16. Optimization and parameterization for multiscale cardiovascular flow simulations using high performance computing – Allison Marsden, University of California – San Diego The treatment and progression of cardiovascular disease is greatly influenced by hemodynamic factors that are often patient-specific. However, standard medical imaging modalities often provide only anatomical information, and cannot be used for design or to predict surgical outcomes. Patient specific simulations offer a means to augment imaging methods and provide predictive tools for medical decision-making. However, major roadblocks to adoption of current simulation methods in the clinic include a lack of cyberinfrastructure that can achieve clinically relevant time frames, and


a lack of tools for efficient manipulation and optimization of surgical designs. In this CAREER award, we are developing a suite of computing tools for high-fidelity parallel simulation of cardiovascular flows and post-operative prediction, as well as robust optimization of surgeries and medical devices. These tools include new physics-based tools for patient specific geometry parameterization, derivative-free optimization methods, and uncertainty quantification in a parallel environment. Our optimization and uncertainty framework results in a multi-layered parallel computing structure, in which multiple cost function evaluations are performed simultaneously, each requiring a multi-processor finite element simulation. Additionally, multiscale simulations require new algorithms to couple 3D computational fluid dynamics simulations with lower order reduced models, to model the circulatory system as a closed loop and predict post-operative flow conditions. These computational tools are being applied to several cardiovascular shape optimization applications, including optimization of graft designs for the Fontan surgery, quantification of hemodynamics in coronary aneurysms caused by Kawasaki disease, and robust design for coronary artery bypass graft surgery. These tools will have broader use in a range of engineering applications requiring coupling between optimization and large scale numerical solvers, including turbulence, combustion, fluid structure interaction, and medical device design. The project also includes an integrated interdisciplinary education and outreach plan that will draw high school students, particularly women and minorities, to the field of engineering and computational science. Our education plan will address training needs in a new interdisciplinary area by exposing students to cardiovascular medicine, and doctors to quantitative simulation-based tools.

17. CAREER: Investigating Fundamental Problems for Underwater Multimedia Communication with Application to Ocean Exploration - Dario Pompili, Rutgers University – New Brunswick The goal of this project is 1) to enable near-real-time acquisition and processing of high-resolution, high-quality, heterogeneous data from mobile and static ocean sensing platforms so to monitor dynamic oceanographic phenomena (e.g., phytoplankton growth and rate of photosynthesis, salinity and temperature gradient, concentration of pollutants); and 2) to enhance the NSF's Ocean Observatories Initiative (OOI) cyberinfrastructure and integrate the OOI and the National Oceanic and Atmospheric Administration (NOAA) ocean observation systems.

18. Distributed Storage Systems for Extreme-Scale Data-Intensive Computing - Ioan Raicu, Illinois Institute of Technology State-of-the-art yet decades-old architecture of HPC storage systems has segregated compute and storage resources, bringing unprecedented inefficiencies and bottlenecks at petascale levels and beyond. This work presents FusionFS and ZHT, two new distributed storage systems designed from the ground up for high scalability (8K-nodes) while achieving significantly higher I/O performance (1TB/sec) and operations per


second (18M/sec). FusionFS and ZHT will achieve these levels of scalability and performance through complete decentralization, and the co-location of storage and compute resources. FusionFS currently supports POSIX-like interfaces important for ease of adoption and backwards compatibility with legacy applications. ZHT has a simple yet functional NoSQL key/value datastore interface to remove unneeded overheads and limitations inherited from POSIX. Both systems are made reliable through data replication with strong and weak consistency semantics, while FusionFS also supports information dispersal algorithms. FusionFS supports scalable data provenance capture and querying, a much needed feature in large scale scientific computing systems towards achieving reproducible and verifiable experiments. Both systems have been deployed on a variety of testbeds, ranging from a 32-node (256-cores) Linux cluster, to a 96-VM virtual cluster on the Amazon EC2 cloud, to a 8K-node (32K-cores) IBM BlueGene/P supercomputer with promising results, when compared to other leading distributed storage systems such as GPFS, PVFS, HDFS, S3, Casandra, Memcached, and DynamoDB. The long term goals of FusionFS and ZHT are to scale them to exascale levels with millions of nodes, billions of cores, petabytes per second I/O rates, and billions of operations per second.

19. Statistical Machine Learning and Big-p, Big-n, Complex Data - Pradeep Ravikumar, University of Texas at Austin Over the past decade, faced with modern data settings, off-the-shelf statistical machine learning methods are frequently proving insufficient. These modern settings pose three key challenges, which largely come under the rubric of "Big Data": (a) the data might have a large number of features, in what we will call "Big-p" data, to denote the fact that the dimension "p" of the data is large, or (b) the data might have a large number of data instances, in what we will call "Big-n" data, to denote the fact that the number of samples "n" is large, or (c) the data-types could be complex: such as permutations, or strings, or graphs, which typically lie in some large discrete space. Interestingly, recent advances in the state of the art methods within these two frameworks share a commonality, based on a simple idea: a complex model parameter could be expressed as a superposition of simple components, which can then hopefully be leveraged for tractable inference and learning. For instance, in graphical model inference, the graph structured parameter is split into simpler graph structured parameters, while in high-dimensional statistics the simpler components are based on particular structure, for instance in compressed sensing, the sparse parameter can be split into a small number of coordinate vectors, and so on. The goal of this CAREER project is to develop a unified framework for statistical machine learning for Big-p, Big-n, Complex Data by leveraging these state of the art developments in graphical models, and high-dimensional statistical methods.


20. Parallel Dynamic Meshing Algorithms, Theory, and Software for Patient-Specific Medical Interventions - Suzanne Shontz, Mississippi State University The overall goal of this NSF CAREER project is to research and develop parallel dynamic meshing algorithms, theory, and software for use in the design of patient-specific medical interventions, such as improved prevention of pulmonary embolism and treatment of hydrocephalus. Pulmonary embolism is a blockage in one or more arteries of the lungs usually caused by a blood clot which has broken free from a deep vein, such as the inferior vena cava (IVC), and traveled to the lungs. Hydrocephalus is a neurological disease in which excess fluid accumulates in the brain, causing the ventricles of the brain to swell, and often causing brain damage. Current simulations of blood clot entrapment by IVC filters and evolution of the brain ventricles in hydrocephalus treatment are of low accuracy. We are developing novel, parallel dynamic meshing algorithms to generate anatomically accurate patient-specific meshes which arise in such simulations. Our parallel dynamic meshing algorithms are used for mesh warping and mesh quality improvement problems that arise based on the patients and medical devices in these simulations. The algorithms will be encapsulated in the form of a parallel dynamic meshing toolkit for use in simulation-assisted medical interventions, as well as numerous other applications, and is being developed for use on a distributed-memory parallel architecture. In addition, we are developing a theoretical framework for dynamic meshing for improved quantitative understanding of deformations. Furthermore, this project integrates research and education via the development of education and outreach activities in computational science and engineering. These activities include the development of a graduate-level course in mesh generation, which incorporates bioengineering applications, and a senior- and graduate-level course in mathematical modeling, computational simulation, and scientific visualization for scientific applications.

21. Multi-Level Performance Modeling for Heterogeneous PetaScale Systems and Beyond – Melissa Smith, Clemson University Future ExaScale computing systems are expected to exceed a hundred-thousand nodes and contain a rich mix of heterogeneous computing resources created by coupling multi-core processors with accelerator processors. However, current applications are not yet able to exploit more than a few tens of thousands of these nodes, or effectively manage this heterogeneous mix of resources (CPUs, GPUs, FPGAs, etc.) to balance workloads. One of the major challenges faced in these systems is user-friendly and accurate heterogeneous performance modeling. Although several performance models exist to fine-tune applications, they are seldom easy-to-use and do not address multiple levels of design space abstraction. Our research aims to bridge the gap between reliable performance model selection and user-friendly analysis. We propose a multi-level performance modeling suite for heterogeneous systems that primarily targets Synchronous Iterative Algorithms (SIAs) using our Synchronous Iterative GPGPU Execution (SIGE) model. The modeling suite addresses different levels of design space


abstraction, thereby enabling developers to choose an optimal performance model that best fits their design goals.

22. Towards an Optimization-Based and Experimentally Verified Predictive Theory of Human Locomotion –Manoj Srinivasan, Ohio State University The research objective of this Faculty Early Career Development (CAREER) award is to obtain a sufficiently accurate, and broadly applicable, predictive theory of how people walk, run, and stabilize their movements. The central hypothesis to be explored and tested systematically is that people move in a manner that minimizes some objective function, perhaps largely correlated with energy use. Towards this end, we will collect new locomotion data and consolidate existing locomotion data under a variety of situations: different speeds, inclines, body changes like addition of masses. This data will be used in a large-scale forward/inverse optimization framework to infer what, if anything, people are minimizing. An accurate and broadly applicable theory of human locomotion will, for instance, enable the systematic model-based design of prosthetic and orthotic devices, and inform the diagnosis and treatment of movement pathologies, including possibly guiding surgical interventions. In the educational component, the PI will work with OSU's Women in Engineering program to create an educational module for female high school students, targeted at increasing their likelihood of pursuing further education in the STEM disciplines. These educational modules will emphasize the interplay between mathematical theories, computer simulations, and experiments in science and engineering, and in improving the quality of human life.

23. LES of full-depth Langmuir circulation in a tidal current - Andres E. Tejada-Martinez, University of South Florida We report on the impact of a tidal current on full-depth Langmuir circulation (LC) in shallow water computed via large-eddy simulations (LES). LC consists of parallel counter rotating vortices that are aligned roughly in the direction of the wind and are generated by the interaction of the wind-driven shear current with the Stokes drift velocity induced by surface gravity waves. During times of weak tidal current, full-depth LC disrupts the classical log-layer dynamics occurring at the bottom of the water column. For example, in terms of mean velocity, the mixing due to LC induces a large wake region eroding the classical log-law profile within the range 90 < z+ < 200. However, during times of strong tidal current, bottom-generated turbulence induced by the tide is able to break-up the full-depth LC giving rise to smaller scale LC characterized by different turbulent structure. The LC turbulent structure during strong and weak tidal currents agrees well with field measurements during episodes of full-depth LC. Furthermore, these results have important implications on turbulence parameterizations for Reynolds-averaged Navier-Stokes simulations of the coastal ocean as these parameterizations often rely on the assumption of log-layer dynamics. The present simulations are representative of a shallow shelf coastal ocean region approximately 15 meters deep and 100 meters by 100 meters in the horizontal. We


describe initial attempts at modifying the LES parallel code structure which will enable LES on wider domains thereby allowing studies of the turbulence on scales of over 1 km by 1km.

24. How much time, energy, and power will my algorithm require? – Richard Vuduc, Georgia Institute of Technology This poster presents a model of algorithms that helps to reason not just about the time needed to execute a computation, but also its physical energy and power. Emerging and future systems, from mobile platforms to data centers and supercomputers, are increasingly limited by these constraints. The poster explores the implications of this model for algorithm design, software engineering, and systems.

25. Enabling Data-Intensive HPC Analytics for interdisciplinary community - Jun Wang, University of Central Florida Today scientists are conjoining two key aspects of computing which will cause a great influx in the oceans of data generated from scientific computing. These two tenets, raw experimental data and computation-rich high-resolution simulations, when combined, have the ability to revolutionize all fields of science, from biology to astrophysics. Dr. Jun Wang's research group has developed a new computing paradigm â€” data-intensive HPC analytics for scientists and engineers to better handle this onslaught of data towards scientific inquiry. Our new interdisciplinary research framework is known as one new data-intensive HPC analytics platform, enabling many scientists and engineers to conduct their big data analyses with complex access patterns in both a super faster and easier way compared with the state-of-the-art solutions. We estimate this could not only save millions of dollars of physicistsâ€™ labor in Los Alamos National Lab, but also significantly shorten the development cycle of analyses programs. Recently, we leverage such data-intensive HPC analytics in Clouds.

26. Scalable Communications - Eric Wang, University of Wyoming One project is about a scalable parallel LSQR (called SPLSQR), a widely used large scale linear system solver. SPLSQR utilizes the particular nonzero structure of matrices to reduce global data communication. Our approach has much more scalable communication volumes with a bounded number of communication neighbors regardless of core counts. Experiments show that SPLSQR are 33 times faster than the PETSc library on around 20K cores. Another project is a performance modeling and optimization analysis tool to predict and optimize the performance of sparse matrix-vector multiplication (SpMV) on GPUs. Based on the performance modeling, we design an SpMV optimal solution auto-selection algorithm to automatically report an optimal solution (i.e., optimal storage strategy, storage format(s), and execution time) for a target sparse matrix. In our experiments,


the average performance improvement is around 44% compared to NVIDIA’s SpMV kernels.

27. CyberGIS for Enabling Data-Intensive Geospatial Discovery and Innovation – Shaowen Wang, University of Illinois at Urbana-Champaign To resolve tremendous challenges of big spatial data that permeate broad scientific and societal realms, GIS needs to be significantly advanced through synergistic integration of computational and geospatial approaches enabled by advanced cyberinfrastructure. CyberGIS - defined as geographic information system (GIS) based on advanced cyberinfrastructure – has emerged as a fundamentally new GIS modality comprising a seamless blending of advanced cyberinfrastructure, GIS, and spatial analysis and modeling capabilities and, thus, has begun to empower scientific breakthroughs and show broad societal impacts. This poster describes major progress, trends, and impacts of cyberGIS for data-intensive geospatial discovery and innovation.

28. Stochastic Multiple Time-Scale Co-Optimized Resource Planning of Future Power Systems with Renewable Generation, Demand Response, and Energy Storage - Lei Wu, Clarkson University This CAREER proposal will develop co-optimized generation, transmission, and DR planning solutions to cope with the impacts of short-term variability and uncertainty of renewable generation (RG) and demand response (DR) as well as hourly chronological operation details of energy storage (ES) and generators. The interaction among variability, uncertainty, and constraints from long-term planning and hourly chronological operation needs be quantified for enhancing security and sustainability of power systems with significant RG, DR, and ES. This project allows better treatment of investment options which require transmission and generation together, in order to exploit favorable sites for wind or solar. The research and educational findings would help educate engineers to meet challenges of the secure and sustainable electricity infrastructure. The project will increase public awareness and understanding of the complexity of power system planning, and appeal to researchers and educators with interests in power systems-based research and education.

29. Multiscale Sensing and Simulations for Bridge Scour – Bill Yu, Case Western Reserve University Bridge scour, which describes the process of soil erosion around bridge foundation, is a major cause of bridge failures. Existing design methods for bridge scour prediction are based on laboratory experimental data and simplified 1D or 2D computational models. These lead to significant errors in bridge scour depth estimation, which either compromises the safety of bridge or makes the design to be overly conservative. This seminar presentation will introduce the recent progresses in my research group to


advance bridge scour research and practice from both computational modeling and innovative sensing aspects. In the sensing aspect, bio-turbulent flow sensor is being developed to help understanding the effects of flow turbulence on soil erosion. An innovative Time Domain Reflectometry sensor is designed for structural health monitoring of bridge scour under the field conditions for decision support. In the modeling aspect, advanced 3D computational fluid dynamics (CFD) models are developed to study the influence of pier shapes on the vortex structures as well as to study the attack angle on flow fields. The final goal of this research is to integrate advanced sensors and computational simulations to build a holistic framework for bridge scour depth prediction and bridge failure risk management.

30. Multi-core CPU and GPU-accelerated Multiscale Modeling for Bimolecular Complexes - Jessica Zhang, Carnegie Mellon University Multi-scale modeling plays an important role in understanding the structure and biological functionalities of large bio molecular complexes. In this paper, we present an ecient computational framework to construct multi-scale models from atomic resolution data in the Protein Data Bank (PDB), which is accelerated by multi-core CPU and programmable Graphics Processing Units (GPU). A multi-level summation of Gaussian kernel functions is employed to generate implicit models for biomolecules. The coecients in the summation are designed as functions of the structure indices, which specify the structures at a certain level and enable a local resolution control on the bio molecular surface. A method called neighboring search is adopted to locate the grid points close to the expected bio molecular surface, and reduce the number of grids to be analyzed. For a speci c grid point, a KD-tree or bounding volume hierarchy is applied to search for the atoms contributing to its density computation, and faraway atoms are ignored due to the decay of Gaussian kernel functions. In addition, multi-core CPU and GPU-accelerated dual contouring mesh generation and quality improvement are employed to generate high quality adaptive tetrahedral meshes from the Gaussian density map. We have applied our algorithm to several large proteins and obtained good results.

31. A Data-Driven Uncertainty-Guided Architecture for Energy Management in Sensor Systems - Alberto Cerpa, University of California-Merced The dramatic growth of wireless sensor networks (WSNs) in the last few years has seen the emergence of a variety of applications and pilot deployments that span scientific, engineering, medical and other disciplines. This project addresses an important research problem in sensor networks: energy management. By using hierarchical WSNs with different resources at different proxy and sensor tiers, as well as novel machine learning methods, we plan to develop DURACEM, a data-driven uncertainty-guided architecture that addresses energy efficiency across many sensor network services in a unified manner. Our architecture is built around (a) prediction models of phenomena, energy usage, and communication and the uncertainty in these models, (b) a slew of services


that can exploit the uncertainty measures to save energy and (c) an overall energy optimization framework that combines the different models and services at both the proxy and sensor tiers.


Speaker Biographies

Farnam Jahanian, Assistant Director for CISE, National Science Foundation Dr. Farnam Jahanian serves as the National Science Foundation Assistant Director for the Computer and Information Science and Engineering (CISE) Directorate. He guides CISE in its mission to uphold the nation's leadership in scientific discovery and engineering innovation through its support of fundamental research in computer and information science and engineering and transformative advances in cyberinfrastructure. Dr. Jahanian oversees the

CISE budget of over $850 million, directing programs and initiatives that support ambitious long-term research and innovation, foster broad interdisciplinary collaborations, and contribute to the development of a computing and information technology workforce with skills essential to success in the increasingly competitive, global market. He also serves as co-chair of the Networking and Information Technology Research and Development (NITRD) Subcommittee of the National Science and Technology Council Committee on Technology, providing overall coordination for the activities of 14 government agencies. Dr. Jahanian holds the Edward S. Davidson Collegiate Professorship in Electrical Engineering and Computer Science at the University of Michigan, where he served as Department Chair for Computer Science and Engineering from 2007 - 2011 and as Director of the Software Systems Laboratory from 1997 - 2000. Earlier in his career, he held research and management positions at the IBM T.J. Watson Research Center. Over the last two decades at the University of Michigan, Dr. Jahanian led several large-scale research projects that studied the growth and scalability of the Internet infrastructure, which ultimately transformed how cyber threats are addressed by Internet Service Providers. His research on Internet infrastructure security formed the basis for the successful Internet security services company Arbor Networks, which he co-founded in 2001. Dr. Jahanian served as Chairman of Arbor Networks until its acquisition in 2010. Dr. Jahanian is the author of over 100 published research papers and has served on dozens of national advisory boards and panels. His work on Internet routing stability and convergence has been highly influential within both the network research and the Internet operational communities and was recognized with an ACM SIGCOMM Test of Time Award in 2008. He has received numerous other awards for his innovative research, commitment to education, and technology commercialization activities. He was named Distinguished University Innovator at the University of Michigan (2009) and received the Governor's University Award for Commercialization Excellence (2005). Dr. Jahanian holds a master's degree and a Ph.D. in Computer Science from the University of Texas at Austin. He is a Fellow of the Association for Computing Machinery (ACM), the Institute


of Electrical and Electronic Engineers (IEEE), and the American Association for the Advancement of Science (AAAS).

Alan Blatecky, Division Director, Advanced Cyberinfrastructure, National Science Foundation Alan Blatecky is the Director of Advanced Cyberinfrastructure at the National Science Foundation. Before coming to NSF, Alan was the Deputy Director of the Renaissance Computing Institute and has held executive leadership positions at the San Diego Supercomputing Center and the North Carolina Research and Education Network. Alan has focused on supporting research

and development initiatives and programs in advanced high performance networking, computing, software, data, and visualization facilities, including early deployment and operations.

Omar Ghattas, the John A. and Katherine G. Jackson Chair inComputational Geosciences, Professor of Geological Sciences andMechanical Engineering, and Director of the Center for Computational Geosciences in the Institute for Computational Engineering and Sciences at the University of Texas at Austin Dr. Omar Ghattas is the John A. and Katherine G. Jackson Chair in Computational Geosciences, Professor of Geological Sciences and Mechanical Engineering, and Director of the Center for Computational Geosciences in the

Institute for Computational Engineering and Sciences (ICES) at The University of Texas at Austin. He also is a member of the faculty in the Computational Science, Engineering, and Mathematics (CSEM) interdisciplinary PhD program in ICES, serves as Director of the KAUST-UT Austin Academic Excellence Alliance, and holds courtesy appointments in Computer Science, Biomedical Engineering, the Institute for Geophysics, and the Texas Advanced Computing Center. He earned BS, MS, and PhD degrees from Duke University in 1984, 1986, and 1988. He has general research interests in simulation and modeling of complex geophysical, mechanical, and biological systems on supercomputers, with specific interest in inverse problems and associated uncertainty quantification for large-scale systems. His center's current research is aimed at large-scale forward and inverse modeling of whole-earth, plate-boundary-resolving mantle convection; global seismic wave propagation; dynamics of polar ice sheets and their land, atmosphere, and ocean interactions; and subsurface flows, as well as the underlying computational, mathematical, and statistical techniques for making tractable the solution and uncertainty quantification of such complex forward and inverse problems on parallel supercomputers. He received the 1998 Allen Newell Medal for Research Excellence, the 2004/2005 CMU College of Engineering Outstanding Research Prize, the SC2002 Best Technical Paper Award, the 2003 IEEE/ACM Gordon Bell Prize


for Special Accomplishment in Supercomputing, the SC2006 HPC Analytics Challenge Award, and the 2008 TeraGrid Capability Computing Challenge award, and was a finalist for the 2008, 2010, and 2012 Bell Prizes. He has served on the editorial boards or as associate editor of 12 journals, has been co-organizer of 12 conferences and workshops and served on the scientific or program committees of 40 others, has delivered plenary lectures at 23 international conferences, and has been a member or chair of 20 national or international professional committees.

William Gropp, Thomas M. Siebel Chair in Computer Science, Computer Science Department; Director, Parallel Computing Institute; Deputy Director for Research Institute for Advanced Computing Applications and Technologies at University of Illinois – Urbana-Champaign William Gropp received his B.S. in Mathematics from Case Western Reserve University in 1977, a MS in Physics from the University of Washington in 1978, and a Ph.D. in Computer Science from Stanford in 1982. He held the positions

of assistant (1982-1988) and associate (1988-1990) professor in the Computer Science Department at Yale University. In 1990, he joined the Numerical Analysis group at Argonne, where he was a Senior Computer Scientist in the Mathematics and Computer Science Division, a Senior Scientist in the Department of Computer Science at the University of Chicago, and a Senior Fellow in the Argonne-Chicago Computation Institute. From 2000 through 2006, he was also Deputy Director of the Mathematics and Computer Science Division at Argonne. In 2007, he joined the University of Illinois at Urbana-Champaign as the Paul and Cynthia Saylor Professor in the Department of Computer Science. In 2008, he was appointed Deputy Director for Research for the Institute of Advanced Computing Applications and Technologies at the University of Illinois. In 2011, he became the founding Director of the Parallel Computing Institute. In 2013, he was named the Thomas M. Siebel Chair in Computer Science. His research interests are in parallel computing, software for scientific computing, and numerical methods for partial differential equations. He has played a major role in the development of the MPI message-passing standard. He is co-author of the most widely used implementation of MPI, MPICH, and was involved in the MPI Forum as a chapter author for MPI-1, MPI-2, and MPI-3. He has written many books and papers on MPI including "Using MPI" and "Using MPI-2". He is also one of the designers of the PETSc parallel numerical library, and has developed efficient and scalable parallel algorithms for the solution of linear and nonlinear equations. Gropp is a Fellow of ACM, IEEE, and SIAM, and a member of the National Academy of Engineering. He received the Sidney Fernbach Award from the IEEE Computer Society in 2008 and the TCSC Award for Excellence in Scalable Computing in 2010.


Steven I. Gordon, Interim Co-Executive Director, Ohio Supercomputer Center Steven I. Gordon, Ph.D. is Interim Co-Executive Director for the Ohio Supercomputer Center (OSC), as well as director of the Center's Ralph Regula School of Computational Science. Gordon is the founding director of the Ralph Regula School of Computational Science. With funding from the Ohio Board of Regents and several National Science Foundation grants, the School has built

inter-disciplinary, inter-institutional programs for computational science education. Currently fourteen Ohio institutions share an undergraduate minor program in computational science that started in 2007. An associate’s degree program and a certificate program began in 2009. Gordon has also played a significant role in several programs in Science Technology Engineering and Mathematics education for high school and middle school students. Those include the Summer Engineering STEM Academy and the Young Women’s Summer Institute. As Senior Director of Client support, he oversees training programs for OSC users and has led workshops for undergraduate faculty and graduate students both at OSC and as part of the SC08 and SC09 education program. As a professor of City and Regional Planning at The Ohio State University, Gordon teaches courses in geographic information systems and environmental modeling and undertakes research in watershed modeling and management. Gordon's research applies models of storm water runoff and water quality to the analysis of watershed management and the applications of communications technology to distance education. He has authored two books and a number of refereed publications in this area. Gordon graduated cum laude from State University of New York at Buffalo with a bachelor's degree in geography. He earned his master's and doctorate degrees from Columbia University in geography with a specialization in environmental systems.

Chandrajit Bajaj, Director, Center for Computational Visualization, Institute for Computational and Engineering Sciences and Professor of Computer Sciences at University of Texas at Austin Chandrajit Bajaj is the director of the Center for Computational Visualization, in the Institute for Computational and Engineering Sciences (ICES) and a Professor of Computer Sciences at the University of Texas at Austin. Bajaj holds the Computational Applied Mathematics Chair in Visualization. He is also an

affiliate faculty member of Mathematics, Electrical and Bio-medical Engineering, Neurosciences, and a fellow of the Institute of Cell and Molecular Biology. He is on the editorial boards for the International Journal of Computational Geometry and Applications, the ACM Computing Surveys, and the SIAM Journal on Imaging Sciences. He is a fellow of the American Association for the Advancement of Science (AAAS), and a fellow of the Association of Computing Machinery (ACM).

http://www.cs.utexas.edu/~bajaj/cvc/bio_img.jpg


Brian I. Athey, Collegiate Professor and Inaugural Chair of the Department of Computational Medicine and Bioinformatics at the University of Michigan Medical School Brian Athey, Ph.D. is Collegiate Professor and Inaugural Chair of the Department of Computational Medicine and Bioinformatics at the University of Michigan Medical School. He is also a Professor of Psychiatry and of Internal

Medicine. He is the founding Principal Investigator of the NIH National Center for Integrative Biomedical Informatics (NCIBI), one of eight NIH National Biomedical Computing Centers, funded by the National Institute on Drug Abuse (NIDA) and the NIH Common Fund. Brian serves as US Academic lead and Co-CEO of the tranSMART Foundation, a non-profit company founded to coordinate the continued development of the open source tranSMART code base which supports an integrated open data sharing and analytics platform used world-wide to accelerate clinical and translational research. Brian has led the National Library of Medicine (NLM) Next-Generation Internet (NGI) Visible Human Project and the DARPA Virtual Soldier Project. He has been a national leader in the NIH Clinical and Translational Scientists (CTSA) informatics Key Function Committee and U-M CTSA Informatics lead. He has over 100 peer-reviewed scientific publications and proceedings, ranging from bioinformatics, metabolomics, chromatin structure, computational biology, optical imaging, and grid computing. In 2000, Brian was awarded a “Peace Fellowship” from the Federation of American Scientists (FAS) for his work in counter biological terrorism in the 1990s.


Program Director Panel

Evelyn M. Goldfield, Program Director, Chemistry Division, National Science Foundation Education: 1983-1984: Postdoctoral, Harvard-Smithsonian Center for Astrophysics, Cambridge MA 1983: Ph.D., Chemistry, University of Illinois at Chicago 1968: M.S., Philosophy, University of Chicago

1963: B.S. (honors), Mathematics, University of Chicago Professional Experience: Sept 2008- present: Program Director, Chemistry Division, National Science Foundation, Chemical Theory Models and Computational Methods Program, program lead Sept 2008- present: Adjunct Associate Professor, Department of Chemistry, Wayne State University, Detroit Sept 2004 – 2000: Associate Professor (Research), Department of Chemistry Wayne State University, Detroit Sept 1994-1999: Senior Research Scientist, College of Science, Wayne State University Sept 1993-1994: Visiting Scholar, Theoretical Chemistry Group, Argonne National Laboratory Sept 1987-1994: Computational Research Associate/Senior Research Scientist, Manager of Computational Science Group, Cornell Theory Center 1984-1987: Research Associate, Department of Chemistry, Cornell University, Research Interests: Quantum Dynamics: Methods and Applications, Dynamics of Chemical Reactions, Chemical Reactions in Confined Environments, High Performance Computing for Quantum Dynamics, Parallel Computing

Daniel S. Katz, Program Director, Division of Advanced Cyberinfrastructure, National Science Foundation Daniel S. Katz is on leave from his position as a Senior Fellow in the Computation Institute, University of Chicago and Argonne National Laboratory. He is also an affiliate faculty member at the Center for Computation and Technology (CCT) and an Adjunct Associate Professor in the Department of

Electrical and Computer Engineering at Louisiana State University (LSU). He is interested in the development and use of advanced cyberinfrastructure to solve challenging problems at multiple scales. His technical interests include applications, algorithms, fault tolerance, and


programming in parallel and distributed computing. His policy interests include citation and credit mechanisms and practices associated with software and data, organization and community practices for collaboration, and career paths for computing researchers. He received his PhD from Northwestern University in 1994 and worked for Cray Research, Jet Propulsion Laboratory, and LSU prior to moving to Chicago in 2009.

Peter McCartney, Program Director, Division of Biological Infrastructure, National Science Foundation Peter McCartney is the program director in the Division of Biological Infrastructure. He oversees research funded under several programs including Biological Databases and Informatics, National Ecological Observatory Network, Cyberinfrastructure for Environmental Observatories, Field Stations and Marine Laboratories, and Assembling the Tree of Life. Prior to NSF he was

a Research Professor in the Global Institute of Sustainability at Arizona State University where he directed projects related to information systems for environmental and archaeological research; use of metadata for designing automated internet access to data and applications; and workflow processing tools for incorporating multiple models into comprehensive analyses.

Thomas Russell, Senior Staff Associate, Office of the Assistant Director, Mathematical and Physical Sciences, National Science Foundation Thomas F. Russell returned to MPS in December 2012 in the Office of the Assistant Director, after a four-year detail as a senior staff associate in the Office of Integrative Activities (OIA). His primary responsibilities in the MPS OAD include leadership roles in DataWay and in the Cyberinfrastructure Framework for 21st Century Science and Engineering (CIF21) OneNSF

investment, as well as continued major contributions to the Integrated NSF Support Promoting Interdisciplinary Research and Education (INSPIRE) OneNSF investment. Dr. Russell joined OIA in November 2008 to coordinate the planning and execution of NSF's Cyber-Enabled Discovery and Innovation (CDI) program and, drawing upon his insight gained from his CDI experience, to formulate best practices regarding the design and implementation of NSF-wide, interdisciplinary programs. Following the conclusion of CDI in 2011, he became a co-chair of the INSPIRE Working Group and led the effort to formulate and manage the INSPIRE funding opportunities in FY 2012 (CREATIV) and FY 2013 (Track 1, Track 2, and Director's Awards). Prior to his detail to OIA, Dr. Russell was on the MPS staff in the Division of Mathematical Sciences (DMS), where he is a program director for computational mathematics and applied mathematics from 2003 to 2008. His integrative activities during his years in DMS included


leading or participating in the following initiatives: the DMS Vertical Integration of Research and Education (VIGRE, 2005-2007); the DMS-GEO Collaborations in Mathematical Geosciences (CMG, 2003-2008); the Interagency Modeling and Analysis Group (IMAG), which supports modeling and analysis of biomedical systems (2004-present); and the NSF-wide, CDI initiative (2007-2011). Previously, Dr. Russell was a professor in the Department of Mathematics at the University of Colorado at Denver from 1987 to 2003, and served as department chair from 1996 to 2001. From 1980 to 1987, he was a research mathematician at the Petroleum Technology Center of Marathon Oil Company in Littleton, CO. He received his Ph.D. in mathematics from the University of Chicago in 1980 under the supervision of Jim Douglas, Jr. His interdisciplinary professional service includes a term as the chair of the Society for Industrial and Applied Mathematics (SIAM) Activity Group on Geosciences from 1998 to 2000, chair of the organizing committee for the 2001 SIAM Conference on Geosciences, associate editor of Water Resources Research from 2001 to 2004, and membership on the scientific council of the French Research Group for Numerical Simulation and Mathematical Modeling of Underground Nuclear Waste Disposal since 2002. His research interests are in the numerical solution of partial differential equations, particularly with applications to subsurface flows in porous media, including groundwater flow and transport and petroleum reservoir simulation. His current major thrusts include control-volume mixed finite element methods, which compute accurate velocities/fluxes for flow equations in heterogeneous media on distorted meshes; Eulerian-Lagrangian localized adjoint methods, which compute accurate solutions for transport equations, even when advection-dominated; efficient algebraic equation solvers for these methods; and upscaling techniques based on stochastic models and the solution of moment equations.

Barry I. Schneider, Program Director, Division of Advanced Cyberinfrastructure, National Science Foundation Dr. Barry I. Schneider is a Program Director for the National Science Foundation's Office of Cyberinfrastructure, specifically for the eXtreme Digital (XD) program. He received his Bachelors in Chemistry from Brooklyn College, his Masters in Chemistry from Yale University and his PhD in Theoretical

Chemistry from the University of Chicago. Before coming to the NSF, he worked at Los Alamos National Laboratory (LANL) in the Theoretical Division, at GTE Laboratories as a member of the Technical Staff, and since 1992 has held visiting appointments at LANL and at the National Institute of Standards and Testing (NIST).


Faculty Attendees Lorena A. Barba is Associate Professor of Mechanical and Aerospace Engineering at the George Washington University, in Washington DC. She has MSc and PhD degrees in Aeronautics from the California Institute of Technology and BSc and PEng degrees in Mechanical Engineering from Universidad Técnica Federico Santa María in Chile. Previous to joining GW, she was Assistant Professor of Mechanical Engineering at Boston University (2008–2013) and Lecturer/Senior Lecturer of Applied Mathematics at University of Bristol, UK (2004–2008). Her research interests include computational fluid

dynamics, especially immersed boundary methods and particle methods for fluid simulation; fundamental and applied aspects of fluid dynamics, especially flows dominated by vorticity dynamics; fast algorithms, especially the fast multipole method and its applications; and scientific computing on GPU architecture. Prof. Barba is an Amelia Earhart Fellow of the Zonta Foundation (1999), a recipient of the EPSRC First Grant program (UK, 2007), an NVIDIA Academic Partner award recipient (2011), and a recipient of the NSF Faculty Early CAREER award (2012). She was appointed CUDA Fellow by NVIDIA Corporation (2012) and is an internationally recognized leader in computational science and engineering. http://lorenabarba.com Collaboration Areas:

• Building community around fast algorithms of the FMM family and developing benchmarks that may help the community evaluate algorithmic innovations and new implementations.

• Solvers and preconditioners for elliptic PDS where FMM may play a role. • Applications of structure-based energy methods in biomolecular physics, where

bioelectrostatics solvers use FMM as a numerical engine.

Nico Cellinese is an Associate Curator at the Florida Museum of Natural History, University of Florida and a Join Associate Professor in the Department of Biology. She is primarily interested in the systematics, evolution and biogeography of flowering plants. Additionally, part of her research revolves around tool development to facilitate biodiversity data synthesis and analysis. http://cellinese.blogspot.com Collaboration Areas:

• Analytical pipelines • workflow implementation

Diego Donzis is an assistant professor in the Department of Aerospace Engineering at Texas A&M University. He is interested in large scale computing, fluid mechanics, turbulence and turbulent mixing in incompressible and compressible flows. He obtained his PhD at Georgia

http://www.blogger.com/profile/05502820618366076308

http://lorenabarba.com/

http://cellinese.blogspot.com/


Tech, and worked in the University of Maryland and the International Center for Theoretical Physics (Italy) before joining the faculty at Texas A&M. http://aero.tamu.edu/people/faculty/?id=529

Benchun Duan is a faculty member in the Department of Geology and Geophysics at Texas A&M University. He is interested in earthquake physics and computational geophysics. He investigates factors and processes that control large earthquake rupture processes, near-field ground motion and deformation. Finite element method and parallel, high-performance computing are technical aspects in his research. His NSF Career project investigates controls on megathrust earthquakes along the Japan Trench subduction zone. He obtained his PhD at UC, Riverside in 2006.

http://geoweb.tamu.edu/profile/BDuan Collaboration Areas:

• Hybrid MPI/OpenMP parallelization of EQdyna, an explicit finite element method (FEM) code, for dynamic rupture and seismic wave propagation simulations.

• FEM mesh generation of complex geological models, including non-planar fault geometry, topography, and complex subsurface velocity structure.

• Visualization of 3D modeling results. • Visualization of earthquake generation processes in the 3D Earth for course

modules

Adrian Feiguin joined Northeastern University as Assistant Professor in 2012, after spending 3 years as Assistant Professor at the University of Wyoming. His field of expertise is computational condensed matter, focusing on theoretical and computational aspects of low dimensional strongly interacting quantum systems. His main interest is understanding exotic phases of matter of quantum origin. This physics is realized under extreme conditions, such as very low temperatures, high pressure, or high magnetic fields, and low spatial dimensions, and it is mostly governed by the collective behavior of the

electrons inside a solid. Strong interactions between these particles can have some dramatic effects, giving rise to some complex and intriguing phenomena of quantum origin, and new phases of matter. Some of this physics could potentially be exploited for technological applications. Examples worth mentioning are the cases of colossal magnetoresistance in manganites (compounds of manganese) for magnetic recording, and high-temperature superconductors, for electric power transmission and magnets. http://www.northeastern.edu/afeiguin/ Collaboration Areas:

• Quantum information • Quantum chemistry

http://aero.tamu.edu/people/faculty/?id=529

http://geoweb.tamu.edu/profile/BDuan

http://www.northeastern.edu/afeiguin/


Antonino Ferrante received his Ph.D. in Mechanical & Aerospace Engineering in 2004 at the University of California, Irvine. From 2004 to 2009, he was postdoctoral scholar at UC, Irvine and at GALCIT, Caltech. Since 2009, he has been Assistant Professor in the William E. Boeing Dept. of Aeronautics & Astronautics at the University of Washington, Seattle. He is recipient of the NSF CAREER (2011), U.S. National Academy of Sciences ICTAM Travel Award (2012), and Royalty Research Fund Award (2012) from the University of Washington. His research interests are in fluid mechanics, multiphase turbulent flows, high-

speed turbulent flows, and chemically-reacting flows. His research tools are direct numerical simulation, large-eddy simulation, and high-performance computing. For his research, he has been developing codes to run on supercomputers since 1998. http://www.aa.washington.edu/faculty/ferrante/

Collaboration Areas: • Petascale elliptic solvers or fast Poisson solvers/parallel FFTs • Optimized massive MPI communications • Exascale computing: hybrid multicore/GPUs, e.g. by using MPI/OpenACC • Parallel I/O and visualizations of petascale data sets • Code optimization on Blue Waters

Baskar Ganapathysubramanian is an Assistant Professor of Mechanical Engineering and Electrical and Computer Engineering at Iowa State University. His research interests are in multi-scale multi-physics modeling, design of materials and processes using computational techniques, and stochastic analysis. The recent focus of his group is on advanced energy technologies including solar cells, and green buildings. Ganapathysubramanian completed his PhD and MS from Cornell University and holds a BS degree from the Indian Institute of Technology-Madras. http://www3.me.iastate.edu/bglab/

Collaboration Areas: • Parallel adaptive mesh generation • Data-mining

Kai Germaschewski is an Assistant Professor at the University of New Hampshire's Department of Physics and Space Science Center. His work focuses on in large-scale computer simulations of plasmas, with applications to the Earth's space environment and laboratory plasmas. He works with both fluid (extended MHD and kinetic (using particle-in-cell) plasma models. He is interested in modern aspects of computing, like adaptive mesh refinement and implicit time integration, and heterogeneous architectures (GPUs, Intel MIC). http://www.eos.unh.edu/Faculty/kgermaschewski

http://www.aa.washington.edu/faculty/ferrante/

http://www3.me.iastate.edu/bglab/

http://www.eos.unh.edu/Faculty/kgermaschewski


Thomas Hacker is an Associate Professor of Computer and Information Technology at Purdue University and Visiting Professor in the Department of Electrical Engineering and Computer Science at the University of Stavanger in Norway. Dr. Hacker’s research interests center around high-performance computing and networking on the operating system and middleware layers. Recently his research has focused on cloud computing, cyberinfrastructure, scientific workflows, and data-oriented infrastructure. Dr. Hacker is also co-leader for Information Technology for the Network for

Earthquake Engineering Simulation (NEES), which brings together researchers from fourteen universities across the country to share innovations in earthquake research and engineering. http://www2.tech.purdue.edu/cpt/SelfStudy/CPTFacultyVitas/FacultyStaff/DisplayStaffMember.asp?member=tjhacker Collaboration Areas:

• High Performance Computing • Reliability • Large-scale Systems

Dr. Ann Jeffers is an Assistant Professor in the Department of Civil and Environmental Engineering at the University of Michigan. Her research lies at the intersections between the fire sciences and structural engineering disciplines, and specifically seeks to establish novel computational methods that bridge the domains of fire science, heat transfer, and structural mechanics. She currently serves on the ASCE Fire Protection Committee and the SFPE Standards Making Committee on the Predicting the Thermal Performance of Fire Resistive Assemblies. Website: http://www-personal.umich.edu/~jffrs

Collaboration Areas: • Multiphysics simulation • High performance computing • Probabilistic methods • Education research

Shantenu Jha is an Assistant Professor at Rutgers University, and a Visiting Scientist at the School of Informatics (University of Edinburgh) and at University College London. Before moving to Rutgers, he was the lead for Cyberinfrastructure Research and Development at the CCT at Louisiana State University. His research interests lie at the triple point of Applied Computing, Cyberinfrastructure R&D and Computational Science. Shantenu is the lead investigator of the SAGA project (http://www.saga-project.org), which is a community standard and is used to support science and engineering

applications on most major production distributed cyberinfrastructure -- such as US NSF's XSEDE and the European Grid Infrastructure. http://radical.rutgers.edu

http://radical.rutgers.edu/people/shantenu-jha/jha_public03-3/

http://www2.tech.purdue.edu/cpt/SelfStudy/CPTFacultyVitas/FacultyStaff/DisplayStaffMember.asp?member=tjhacker

http://www2.tech.purdue.edu/cpt/SelfStudy/CPTFacultyVitas/FacultyStaff/DisplayStaffMember.asp?member=tjhacker

http://www-personal.umich.edu/~jffrs

http://radical.rutgers.edu/


Collaboration Areas: • Domains of science and engineering that entail distributed and high-

performance computing

Kapil Khandelwal is an Assistant Professor in the Department of Civil & Environmental & Earth Sciences at the University of Notre Dame. He received BS in Civil Engineering from IIT-Roorkee (India), MS in Structural Engineering From IIT-Delhi (India) and Ph.D. in Civil Engineering from the University of Michigan, Ann Arbor. His research interested includes: computational solid mechanics (FEM), gradient elasticity/plasticity, computational fracture mechanics, topology optimization, and progressive collapse of structural systems. http://www3.nd.edu/~kkhandel/

Collaboration Areas: • Multiscale mechanics • Large scale optimization • GPU/parallel algorithms and optimization • Uncertainty quantification

Xiaolin (Andy) Li is an associate professor in Department of Electrical and Computer Engineering at University of Florida. His research interests include Parallel and Distributed Systems, Cyber-Physical Systems, and Network Security & Privacy. He is directing the Scalable Software Systems Laboratory (S3Lab). He is in the executive committee of IEEE Technical Committee of Scalable Computing (TCSC) and the coordinator of BigData & MapReduce and Sensor Networks. He has been a TPC chair for several international conferences and workshops and an associate editor for several journals. He received a PhD in

Computer Engineering from Rutgers University. He is a recipient of the National Science Foundation CAREER Award 2010 and a member of IEEE and ACM. http://www.andyli.ece.ufl.edu/

Laurence Loewe is an Assistant Professor at the University of Wisconsin- Madison. He investigates questions in the new field of evolutionary systems biology, which merges systems biology and population genetics. To enable this, his group is developing two major tools. The first, Evolvix, is a new programming language that makes it easy for biologists to build simulation models linked to real data. The second, Evolution@home, is a globally distributed computing system that is being redesigned for analyzing the flood of simulation data generated by Evolvix models. His group works-in silico-on

diverse topics like circadian clocks, antibiotic resistance evolution, the population genetics of harmful mutations and species extinction. He is interested in bridging the gap between simple analytically understandable mathematical models and biological reality by building rigorous simulation models to answer various evolutionary questions.

http://www3.nd.edu/~kkhandel/

http://www.andyli.ece.ufl.edu/


• 2011 Assistant Professor, University of Wisconsin-Madison • 2007 Postdoc, Center for Systems Biology Edinburgh, University of Edinburgh • 2006 Lecturer in Evolutionary Genetics, Institute of Evolutionary Biology, University

of Edinburgh • 2003 Postdoc, Institute of Evolutionary Biology, University of Edinburgh, UK • 2002 Dr. rer. nat., Technical University Munich, Germany

http://evolution.ws/people/loewe http://evolvix.org

Kamesh Madduri is an assistant professor in the Computer Science and Engineering department at The Pennsylvania State University. He received his PhD in Computer Science from Georgia Institute of Technology's College of Computing in 2008, and was previously a Luis W. Alvarez postdoctoral fellow at Lawrence Berkeley National Laboratory. His research interests include high-performance computing, parallel graph algorithms, and massive scientific data analysis. He is a member of IEEE, ACM, and SIAM. http://www.cse.psu.edu/~madduri/

Collaboration Areas: • Big data analysis and mining applications • Scalable data management • Visualization

Alison Marsden is an associate professor and Jacobs Faculty Fellow in the Mechanical and Aerospace Engineering department at the University of California San Diego. She graduated with a bachelor's degree in mechanical engineering from Princeton University in 1998, a PhD in mechanical engineering from Stanford in 2005, and did postdoctoral work at Stanford University in bioengineering and pediatric cardiology from 2005-07. She has been the recipient of an American Heart Association postdoctoral fellowship, an AHA beginning grant in aid award, a Burroughs Wellcome Fund Career

Award at the Scientific Interface, an NSF CAREER award, and is a member of an international Leducq Foundation Network of Excellence. Her work focuses on the development of numerical methods for simulation of cardiovascular blood flow problems, medical device design, application of optimization to fluid mechanics, and use of engineering tools to impact patient care in cardiovascular surgery and congenital heart disease. http://maeresearch.ucsd.edu/marsden/AMarsden/Home.html Collaboration Areas:

• Image segmentation • Machine learning algorithms • Uncertainty quantification

http://evolution.ws/people/loewe

http://evolvix.org/

http://www.cse.psu.edu/~madduri/

http://maeresearch.ucsd.edu/marsden/AMarsden/Home.html


Dr. Pompili is an Associate Professor at Rutgers University-New Brunswick, where he is the director of the Cyber-Physical Systems Laboratory (CPS Lab), the site co-director of the NSF Center for Cloud and Autonomic Computing (CAC), and the associate director of application collaborations of the Rutgers Discovery Informatics Institute (RDI2). In 2011, Dr. Pompili was awarded the NSF CAREER to design efficient communication solutions for underwater multimedia applications and the Rutgers/ECE Outstanding Young Researcher award. In 2012, he received the ONR Young Investigator Program (YIP) award

to develop an uncertainty-aware autonomic mobile computing grid framework as well as the DARPA Young Faculty Award (YFA) to enable complex real-time information processing based on compute-intensive models for operational neuroscience; that year he was also awarded the Rutgers/ECE Excellence in Research award. His research spans underwater acoustic communication and coordination of vehicles, ad hoc and sensor networks, thermal management of datacenters as well as mobile and green computing. More info at: http://www.ece.rutgers.edu/~pompili/ Collaboration Areas:

• Mobile grid computing • Underwater communications • Coordination of vehicles • Cloud-assisted robotics • Wearable computing

Dr. Ioan Raicu is an assistant professor in the Department of Computer Science (CS) at Illinois Institute of Technology (IIT), as well as a guest research faculty in the Math and Computer Science Division (MCS) at Argonne National Laboratory (ANL). He is also the founder (2011) and director of the Data-Intensive Distributed Systems Laboratory (DataSys) at IIT. He has received the prestigious NSF CAREER award (2011 - 2015) for his innovative work on distributed file systems for exascale computing. He was a NSF/CRA Computation Innovation Fellow at Northwestern University in 2009 - 2010,

and obtained his Ph.D. in Computer Science from University of Chicago under the guidance of Dr. Ian Foster in March 2009. He is a 3-year award winner of the GSRP Fellowship from NASA Ames Research Center. His research work and interests are in the general area of distributed systems. His work focuses on a relatively new paradigm of Many-Task Computing (MTC), which aims to bridge the gap between two predominant paradigms from distributed systems, High-Throughput Computing (HTC) and High-Performance Computing (HPC). His work has focused on defining and exploring both the theory and practical aspects of realizing MTC across a wide range of large-scale distributed systems. He is particularly interested in resource management in large scale distributed systems with a focus on many-task computing, data intensive computing, cloud computing, grid computing, and many-core computing. Over the past decade, he

http://www.ece.rutgers.edu/~pompili/


has co-authored 86 peer reviewed articles, book chapters, books, theses, and dissertations, which received over 3250 citations, with a H-index of 22. His work has been funded by the NASA Ames Research Center, DOE Office of Advanced Scientific Computing Research, the NSF/CRA CIFellows program, and the NSF CAREER program. He has also founded and chaired several workshops, such as ACM Workshop on Many-Task Computing on Grids and Supercomputers (MTAGS), the IEEE Int. Workshop on Data-Intensive Computing in the Clouds (DataCloud), and the ACM Workshop on Scientific Cloud Computing (ScienceCloud). He is on the editorial board of the IEEE Transaction on Cloud Computing (TCC), the Springer Journal of Cloud Computing Advances, Systems and Applications (JoCCASA), as well as a guest editor for the IEEE Transactions on Parallel and Distributed Systems (TPDS), the Scientific Programming Journal (SPJ), and the Journal of Grid Computing (JoGC). He has been leadership roles in several high profile conferences, such as HPDC, CCGrid, Grid, eScience, and ICAC. He is a member of the IEEE and ACM. More information can be found at http://www.cs.iit.edu/~iraicu/, http://datasys.cs.iit.edu/, and http://www.linkedin.com/in/ioanraicu. Website: http://www.cs.iit.edu/~iraicu/ Collaboration Areas:

• Data-Intensive Computing Applications (requiring either POSIX or NoSQL interfaces)

• Communities interested in Data Provenance • Large-scale Workflow-based Applications

Pradeep Ravikumar received his B.Tech. in Computer Science and Engineering from the Indian Institute of Technology, Bombay, and his PhD in Machine Learning from the School of Computer Science at Carnegie Mellon University. He was then a postdoctoral scholar at the Department of Statistics at the University of California, Berkeley. He is now an Assistant Professor in the Department of Computer Science, at the University of Texas at Austin. He is also affiliated with the Division of Statistics and Scientific Computation, and the Institute for Computational Engineering and Sciences at UT Austin. His

thesis has received honorable mentions in the ACM SIGKDD Dissertation award and the CMU School of Computer Science Distinguished Dissertation award. He is also a recipient of the NSF CAREER Award. http://www.cs.utexas.edu/~pradeepr/ Collaboration Areas:

• Applications involving large-scale graphical models • Computational Music/Speech • Time-varying Data Applications

http://www.cs.iit.edu/~iraicu/

http://datasys.cs.iit.edu/

http://www.linkedin.com/in/ioanraicu

http://www.cs.iit.edu/~iraicu/

http://www.cs.utexas.edu/~pradeepr/


Suzanne Shontz is an Assistant Professor in the Department of Mathematics and Statistics at Mississippi State University. She is also affiliated with the Department of Computer Science and Engineering, the Center for Computational Sciences, and the Graduate Program in Computational Engineering at Mississippi State. Suzanne’s research is in parallel scientific computing and focuses on the development of meshing and numerical optimization algorithms and their applications to medicine, image processing, and electronic circuits, to name just a few. Suzanne is the recipient of a 2011

NSF CAREER Award and a 2011 NSF PECASE Award from President Obama for her research in computational- and data-enabled science and engineering. Along with Thomas Hacker of Purdue University, she is a Co-Chair of the 2012 and 2013 NSF CyberBridges Workshops. http://sshontz.math.msstate.edu Collaboration Areas:

• Mathematical modeling • Model order reduction • Scientific visualization • Patient data sets

Before her appointment to Clemson in 2006, Dr. M. Smith was a research associate at the Oak Ridge National Laboratory (ORNL) for 12 years. In 2004, Dr. Smith began collaborations with the newly formed Future Technologies Group at ORNL and conducted research on emerging computing architectures including reconfigurable computers, multi-core, and optical processors. She continues to collaborate with some of the top research scientists at ORNL and across the country in areas of heterogeneous high-performance computing and System Performance Modeling and Analysis.

Dr. Smith’s current research activities focus on the applied use of emerging heterogeneous computing architectures. Her research group is interested in the performance computer architectures for various application domains including scientific applications (modeling and simulation), high-performance or real-time embedded applications, and medical and image processing. Her group explores optimization techniques and performance analysis for emerging heterogeneous platforms, including many processors, Graphical Processing Units (GPUs) and Field-Programmable Gate Array-based (FPGA-based) reconfigurable computers. Also of interest are the tools and methodologies that are needed to efficiently and effectively program and utilize these architectures. http://www.clemson.edu/ces/departments/ece/faculty_staff/faculty/msmith.html Collaboration Areas:

• Big data applications • Heterogeneous HPC system users • Performance modeling and analysis

http://sshontz.math.msstate.edu/

http://www.clemson.edu/ces/departments/ece/faculty_staff/faculty/msmith.html


Manoj Srinivasan is an assistant professor in the Department of Mechanical and Aerospace Engineering at the Ohio State University. His recent research has focused on the understanding of human locomotion and biomechanics from the perspective of optimal control and dynamical systems theory, but he is broadly interested in a variety of biological and mechanical systems. Srinivasan received an undergraduate degree in engineering from the Indian Institute of Technology, Madras, received a doctoral degree in Theoretical and Applied Mechanics at Cornell University, and was a post-doctoral

researcher and lecturer at Princeton University. He is an NSF CAREER award recipient. http://movement.osu.edu A small fraction of our work on my CAREER grant will be devoted to the collection, consolidation, documentation, and dissemination of a certain class of locomotion data. Collaboration Areas:

• Data dissemination • Format standardization • Management

Andrés Tejada-Martínez is associate professor in Civil and Environmental Engineering at University of South Florida. Tejada-Martínez has received an NSF CAREER Award and various others NSF collaborative research awards for his work in large-eddy simulations of turbulent mixing in shallow shelf coastal regions and in the upper ocean mixed layer. www.eng.usf.edu/~aetejada Collaboration Areas:

• Finite difference, finite elements and spectral methods for fluid dynamics • Stratified flows • Parallel computing

Richard (Rich) Vuduc is an Associate Professor at the Georgia Institute of Technology ("Georgia Tech") in the School of Computational Science and Engineering. His research lab, the HPC Garage (hpcgarage.org), is interested in all-things-high-performance-computing, with an emphasis on parallel algorithms, performance analysis, and performance tuning. He is a recipient of the NSF CAREER Award, member of the DARPA Computer Science Study Panel, and co-recipient of the Gordon Bell Prize (2010). His lab's work has

received a number of best paper nominations and awards including most recently the 2012 Best Paper Award from the SIAM Conference on Data Mining. In 2013, he received a Lockheed Martin Excellence in Teaching Award.

http://movement.osu.edu/

http://www.eng.usf.edu/~aetejada


Collaboration Areas: • Scalable applications and libraries • Programming models • Computer architecture

Dr. Jun Wang is an Associate Professor in Department of Electrical Engineering and Computer Science at the University of Central Florida, Orlando, FL, USA. He has conducted extensive research in the areas of Computer Systems and High Performance Computing. His specific research interests include data-intensive high performance computing, massive storage and file system, I/O Architecture, and low-power computing. http://www.eecs.ucf.edu/~jwang

Collaboration Areas: • Data-intensive HPC applications • Software defined network • GPU computing • Memory architecture

Liqiang (Eric) Wang is a Castagne Associate Professor in the Department of Computer Science at the University of Wyoming. He has been an assistant professor (2006-2012) and an associate professor (2012-present) in the same department. He received Ph.D. in Computer Science from Stony Brook University in 2006. His research interest is the design and analysis of parallel systems for big-data computing, which includes two aspects: design and analysis. For design, he is currently working on optimizing performance, scalability, resilience, and load balancing of data-intensive computing,

especially on Cloud, GPU, and multicore platforms. For the aspect of analysis, he focuses on using program analysis to detect programming errors and performance defects in large-scale parallel computing systems. He received an NSF CAREER Award in 2011. http://www.cs.uwyo.edu/~lwang7/ Collaboration Areas:

• Scalability of large-scale linear equation solver • Automatic load balancing of scientific computations • Automatic cloud provisioning for HPC • Storage and I/O optimization for big-data computing

Shaowen Wang is an Associate Professor of Geography and Geographic Information Science (Primary), Computer Science, and Urban and Regional Planning at the University of Illinois at Urbana-Champaign (UIUC), where he was named Helen Corley Petit Scholar for 2011-2012. He is also Associate

http://www.eecs.ucf.edu/~jwang

http://www.cs.uwyo.edu/~lwang7/


Director for CyberGIS and a Senior Research Scientist of the National Center for Supercomputing Applications (NCSA), and Founding Director of the CyberInfrastructure and Geospatial Information Laboratory. He holds affiliate appointments within UIUC’s Computational Science and Engineering Graduate Program and Illinois Informatics Institute. He received his BS in Computer Engineering from Tianjin University in 1995, MS in Geography from Peking University in 1998, and MS of Computer Science and PhD in Geography from the University of Iowa in 2002 and 2004 respectively. His research and teaching interests center on three interrelated themes: 1) computational theories and methods in geographic information science, 2) cyberinfrastructure and data-intensive computational science, and 3) multi-scale geospatial problem solving and spatiotemporal synthesis. He has published a number of peer-reviewed papers including articles in more than 15 journals. He has served as an Action Editor of GeoInformatica, and guest editor or editorial board member for six other journals, book series and proceedings. He served on the University Consortium for Geographic Information Science Board of Directors from 2009 to 2012, and was appointed two terms as a Councilor of the Open Science Grid Consortium. He was a visiting scholar at Lund University sponsored by the National Science Foundation (NSF) in 2006 and NCSA Fellow in 2007, and received the NSF CAREER Award in 2009. Website: http://www.cigi.illinois.edu/shaowen/ Collaboration Areas:

• Advanced cyberinfrastructure • Data-intensive geospatial sciences and technologies • Scalable computing and information systems • Sustainability science

Dr. Lei Wu is an Assistant Professor of ECE Department at Clarkson University. He has experience working with NYISO, GE, and Siemens Energy Automation on various power system studies. He has extensive publications on power systems research and is the recipient of the IEEE Transactions Prize Paper Award from the IEEE Power and Energy Society in 2009. He serves on Research Committee of the Clarkson’s Honors program and is a member of the Center for Sustainable Energy Systems. He is an Editor of IEEE Transactions on Sustainable Energy and Guest Editor of IEEE Transactions on

Smart Grid. His educational and research activities are supported by grants from NSF, DOE, GE, and IBM. http://people.clarkson.edu/~lwu/ Collaboration Areas:

• High performance computing • Data analytics • Mathematical optimization

http://www.cigi.illinois.edu/shaowen/

http://people.clarkson.edu/~lwu/


Dr. Xiong (Bill) Yu is an associate professor at the Department of Civil Engineering, Case Western Reserve University. He also holds courtesy appointments in the Department of Electrical Engineering and Computer Science, the Department of Materials Science and Engineering, CWRU. Dr. Yu received civil engineering training via Ph.D. degree from Purdue University, B.S. and M.S. degrees from Tsinghua University, China. His interdisciplinary training includes M.S. degree from EECS from Purdue University and B.S. degree in computer science from Tsinghua University. His research interest is

in the broad area of geoengineering related to infrastructure sustainability, environment and energy needs. His work embraces innovative sensors and materials to improve sustainability and intelligence of the civil infrastructure systems. He is the PI of over 25 research projects sponsored by the National Science Foundation , National Research Council, Ohio DOT, Federal Highway Administration, NCHRP-IDEA and other agencies such and private industry. Dr. Yu is a member of ASCE, ISSMGE, ASTM, TRB, SPIE and ASNT. He serves on SHRP and NCHRP project panels and as chair of G-I Engineering Geology and Site Characterization committee. He is a member of editorial board of four ASCE and ASTM journals. Dr. Yu is a recipient of the NSF CAREER award in 2009. He has published over 120 papers (including 54 journal papers) and co-edited three GSPs. He holds a few patents and provisional patents, and invention disclosures. He has advised the research of 11 Ph.D. students and graduated 5, among whom three are employed with tenure track faculty positions in the U.S. He is selected to participate in the U.S.-Germany Frontiers of Engineering symposium in 2013 as well as the inaugural Global Frontiers of Engineering symposium in 2013.

http://filer.case.edu/xxy21/Index.html

Jessica Zhang is an Associate Professor in Mechanical Engineering at Carnegie Mellon University with a courtesy appointment in Biomedical Engineering. Her research interests include computational geometry, mesh generation, computer graphics, visualization, finite element method, isogeometric analysis and their application in computational biomedicine and engineering. She is the recipient of a 2012 NSF CAREER Award. http://www.andrew.cmu.edu/user/jessicaz/

Collaboration Areas: • Computational biology • Finite element applications

http://filer.case.edu/xxy21/Index.html

http://www.andrew.cmu.edu/user/jessicaz/


Alberto Cerpa was one of the three founding faculties of the Electrical Engineering and Computer Science program in the School of Engineering at UC Merced when he joined in 2005. He received a Ph.D. degree in Computer Science from UCLA (2005), working under the supervision of Deborah Estrin. He also received a M.Sc. in Computer Science from USC (2000), and a M.Sc. in Electrical Engineering from USC (1998). Alberto received his undergraduate degree (Engineer) in Electrical Engineering from Buenos Aires Institute of Technology, in

Buenos Aires, Argentina (1995). His interests lie broadly in the computer networking and distributed systems areas, with recent focus in systems research in wireless sensor networks. Alberto is a recipient of the NSF CAREER Award (2013). Several of his papers are some of the top cited papers in top journals, including ACM TMC (2nd out of 1535), ACM SIGCOMM CCR (32 out of 2472), and JPDC (59 out of 2867). He is a member of the ACM and IEEE. http://www.andes.ucmerced.edu/~acerpa/ Collaboration Areas:

• Optimization algorithms, specifically those that can be implemented in very resource constraint environments (extremely limited computation and storage)

• Control algorithms for building HVAC control, e.g. Model Predictive Control (MPC)

• Machine Learning, in particular novel techniques for building predictions models of phenomena.

http://www.andes.ucmerced.edu/~acerpa/

Workshop Program - Center for Computational Sciences

Documents

Transcript of Workshop Program - Center for Computational Sciences