Joel E. Tohline - US grants

Tohline 95-28424 Dr. Tohline will study stellar formation using the gravitational hydrodynamics code developed previously. Specifically, he will address three problems -- to map out the critical states at which protostellar accretion disks become dynamically unstable to gravitational fragmentation, to examine the dynamical stability of ellipsoidal or dumbbell-shaped equilibrium structures with nontrivial internal motions, and to examine the evolution of warped HI galaxy disks to determine the extent to which the disk settling process is connected to the global process of star formation. ***

High performance computing (HPC) is a major initiative at Louisiana State University (LSU). Its multidisciplinary research programs focus on grand challenges in materials science, astronomy, microsystems design, and environmental studies. To advance this effort, we request five NSF Traineeships. Consortial arrangements have been made to recruit in-state and out-of-state students, especially minorities from Historically Black Institutions. Innovative training structures have been introduced to integrate HPC into graduate education and research. These include new cross-disciplinary curricula and a dual-degree program to allow Ph.D. students in the physical sciences to get a M.S. from Computer Science and vice versa. Unique opportunities are available for graduate trainees to interact with scientists at Oak Ridge and Argonne National Laboratories, Mobil and Ethyl Corporations, and in Europe and Japan. To establish the HPC effort a LSU, two excellent parallel computing laboratories have been set up with $2. million in infrastructure grants from the State of Louisiana. These laboratories feature an 8,192-node MasPar, a 64-cell iWarp, and 8-node iPSC/860, and an 8-processor Silicon Graphics. With the latest grant of $850,000, we plan to acquire a 56-node Intel Paragon.

Bodenheimer, Peter
"Evolution of Protostellar Planets"
AST-9987417

Dr. Bodenheimer continues his investigation of the development and evolution of protostellar condensations and disks. Particular emphasis will be placed on the development of protoplanets and their role in moderating the later stages of protostellar disks. The results will directly contribute to a better understanding of how the solar system formed, and the most likely formation scenarios for the wide assortment of newly discovered extrasolar planets.
Funding for this project was provided by the NSF program for Stellar Astronomy & Astrophysics (AST/SAA).

This ITR-Medium project will combine expertise in applied mathematics, numerical relativity, computer science, and astrophysics to develop techniques needed to simulate realistic strong-gravitational-field astrophysical systems. To this end, a driver-problem is chosen that involves solving the equations of Einstein's theory of general relativity coupled to electromagnetism and hydrodynamics to describe phenomena such as black hole formation and the accretion of matter around a black hole. Such efforts have historically encountered a number of difficulties, and therefore a number of advanced techniques will be brought to bear upon these problems.
Maintaining stability in such evolutions will be studied using properties of hyperbolic equations, and numerical techniques will be developed to take advantage of useful analytical properties. At the same time, development of distributed adaptive mesh refinement will allow for the most efficient use of computational power.

A wide range of physical phenomena are modeled by these and similar equations, and thus the techniques and developments studied here should be applicable to a wide range of problems. The physical situations under study attract much interest, and better models of their dynamics will
benefit, in particular, the analysis involved in the LIGO project to detect gravitational waves. This project also will train scientists in these methods as well as publicize them with papers and conferences.

Rotating stars are susceptible to a variety of global, gravitationally driven instabilities, especially during the earliest (e.g., protostellar) and latest (e.g., compact object) phases of their evolution when they are most likely to be rotating rapidly. These include, for example, a wide spectrum of nonradial pulsations. An even larger number of instabilities may be encountered by stars in binary systems through their mutual gravitational interaction, in particular, tidal and mass transferring instabilities. Any one of these instabilities may result in substantial alterations in the star's global structure and change the star's evolution. In the case of compact objects, such as neutron stars, they may give rise to measurable levels of gravitational radiation. Also, we're not likely to understand the origin of binary stars until we are able to realistically model the variety of gravitationally driven instabilities in rapidly rotating protostars and protostellar gas clouds. The Principle Investigator of this project is extending his detailed investigations of gravitationally driven instabilities in rotating stars, using fully three-dimensional, nonlinear hydrodynamical techniques to determine under what conditions certain key instabilities arise, and what the astrophysically important implications are of the nonlinear development of these instabilities.

Broader Impacts. The types of high-performance computational tools that must be developed in order to further our understanding of gravitational instabilities in astrophysical systems are also used to make significant research advancements in atmospheric sciences, coastal studies, certain areas of the health sciences, and numerous subfields of engineering. This Louisiana State University researcher and his students regularly exchange ideas and computational techniques with applied mathematicians, computer scientists, and numerous colleagues who employ computational fluid dynamics tools in their research to model hurricane storm surges, fluid flow in the human eye, cooling flows in advanced gas turbine systems, coastal erosion, and reacting flows in petrochemical stirring tanks. Most of these areas of research directly influence Louisiana's environmental and economic well-being. The Principle Investigator will continue to be involved in K-12 and general public outreach projects, especially those associated with the local Highland Road Park Observatory and one secondary school in Baton Rouge where substantial efforts are being made to strengthen the early training of young women in science and mathematics.

A multidisciplinary graduate program of education, research and training in Multi-Scale Computations of Fluid Dynamics (CFD) at Louisiana State University (LSU) will be undertaken in interdisciplinary partnership between the various CFD groups and the Center for Computation and Technology at LSU, and outreach partners at Southern University, Louisiana Tech University, and LSU Eye Center. All schools are tightly connected by a 40 Gbit optical network and tied to the National LambdaRail. The intellectual merit and purpose of this program is to provide doctoral students with enhanced multidisciplinary education and training that will integrate all elements critical in solving critical CFD projects of the future: distributed collaborations connected by optical networks, high performance and grid computing techniques, CFD as a fundamental discipline, and numerous fluid dynamical application areas where Louisiana has unique research strengths. Braoder impacts of the project relate to application areas that span the spectrum of flow scales (from microns to kilometers) and include biological/biomedical flows, estuarine/oceanic flows, reservoir flows, and astrophysical flows. IGERT research and education will occur at the disciplinary interfaces, with faculty mentors from two or more disciplines, and a focus on enabling large-scale parallel computing of flow systems that resolve scales and their dynamics that were previously not possible. IGERT students will complete a program of study that includes an original interdisciplinary research problem for their dissertations, and a mix of interdisciplinary fluid dynamics, computational science and CFD courses that are team-taught and cross-listed across the various departments. IGERT is an NSF-wide program intended to meet the challenges of educating U.S. Ph.D. scientists and engineers with the interdisciplinary background, deep knowledge in a chosen discipline, and the technical, professional, and personal skills needed for the career demands of the future. The program is intended to catalyze a cultural change in graduate education by establishing innovative new models for graduate education and training in a fertile environment for collaborative research that transcends traditional disciplinary boundaries.

AST-0708551
Tohline

Rotating stars are susceptible to a variety of global, gravitationally driven instabilities, especially during the earliest and latest phases of their evolution. Stars in binary systems may encounter an even larger number of instabilities through their mutual gravitational interaction. Once activated, any one of these instabilities can be amplified enough to drive a substantial alteration in the star's global structure, altering its evolution and perhaps emitting measurable gravitational radiation. In fact, the ability to detect gravitational radiation will depend on the accurate prediction of the expected spectrum. This project will extend existing work on gravitationally driven instabilities in rotating stars, using fully three-dimensional, nonlinear hydrodynamical techniques to determine the conditions for certain key instabilities, and to study the astrophysically important implications of the nonlinear development of instabilities.

The required computational algorithms and hardware tools also have significant research applications in atmospheric sciences, coastal studies, health sciences, and engineering. Professor Tohline and his group regularly exchange ideas and computational techniques with applied mathematicians, computer scientists, and numerous colleagues who employ computational fluid dynamics in their research to model hurricane storm surges, fluid flow in the human eye, cooling flows in advanced gas turbine systems, coastal erosion, and reacting flows in petrochemical stirring tanks. Most of these areas of research directly influence Louisiana's environmental and economic wellbeing. The research team will also continue their K-12 and general public outreach projects, especially those associated with a local observatory, and with a secondary school where substantial efforts are being made to strengthen the early training of young women in science and mathematics.

An accreting black hole binary system (ABHB) consists of a star orbiting a small black hole. One particularly interesting example is a neutron star-black hole (NS-BH) binary, which may be interesting sources of gravitational waves for gravitational wave observatories such as Advanced LIGO. The gravitational and electromagnetic wave signals from NS-BH binaries contains information about the particular evolution of the binary, including magnetic effects and the disruption of the neutron star by the tidal forces of the black hole.

This award supports the development of computational infrastructure and techniques to study ABHBs and related systems. This includes the further development of HAD, a freely available, modular computational toolkit for solving differential equations in a distributed parallel environment. HAD provides sophisticated adaptive mesh refinement capabilities, which allow the user to combine different types of differential equations and/or numerical methods. Techniques for solving the Einstein equations and magnetohydrodynamics (MHD) equations will be further developed and implemented in the code to meet the demands of simulating ABHB systems and extracting valuable physical information.

This work serves to advance numerical techniques for an important scientific problem, as well as the creation and dissemination of advanced tools for distributed computing. The modular design of the infrastructure allows other researchers to share and extend these tools. Finally, this study involves the training of postdoctoral researchers, graduate students, and undergraduates.

The work to be carried out under this award will develop the needed infrastructure and apply it to study realistic compact binary systems that can give rise to gamma ray burst phenomena. To this end the required physics modules will be incorporated in a new infrastructure framework able to exploit the new generation of high performance computers. The implementations of this infrastructure will allow study of the system in depth, with the goal of predicting and interpreting aspects of observations of electromagnetic and gravitational waves.

This work serves to advance computational and numerical techniques for an exciting scientific problem, as well as the creation and dissemination of advanced tools for distributed computing. The research to be carried out has as its goal to give an unprecedented description of possible (short) gamma ray burst systems and to develop a new computational model to efficiently utilize petascale computing. Finally, this study involves the training of postdoctoral researchers, graduate students, and undergraduates.

Scaling scientific problems to 10,000,000 processors for the next generation HEC systems is today severely challenged by conventional practices of programming models, languages, and their supporting compilation systems. To achieve this goal one must expose greater degree of parallelism and improve parallel computing efficiency than is otherwise feasible with conventional methods such as MPI.
The goal of this collaborative research project is to dramatically enhance the scalability of challenging physics problems, through the application of an innovative programming model. The strategy is to replace static message-passing course grained processes using global barrier synchronization in a distributed memory space with a model using dynamic message-driven multiple threads using lightweight synchronization objects in a partitioned global address space. Parallelism is to be extracted directly from the large irregular sparse and time varying data structures. Ephemeral user-threads will permit many simultaneous tasks over the data structures, exposing the intrinsic near-fine grain parallelism. System-wide latency will be hidden by overlapping computation with communication through the advanced communication strategy of asynchronous message-driven processing. Consequently this will enable a class of physics problems that cannot currently be done using conventional methods.

This proposal will be awarded using funds made available by the American Recovery and Reinvestment Act of 2009 (Public Law 111-5), and meets the requirements established in Section 2 of the White House Memorandum entitled, Ensuring Responsible Spending of Recovery Act Funds, dated March 20, 2009.

The STCI Development of Stork Data Scheduler for Mitigating the Data Bottleneck in Petascale Distributed Computing Systems project will enhance the Stork data scheduler to mitigate the end-to-end data handling bottleneck in petascale distributed computing systems and make it available for a wide range of user community as production quality software. New functionalities of Stork will include: data aggregation and caching; early error detection, classification, and recovery; integration with workflow planning and management; optimal protocol tuning; and data transfer performance prediction services.

Intellectual Merit:
The Stork data scheduler will make a distinctive contribution to petascale distributed computing in the areas of planning, scheduling, monitoring and management of data placement tasks and application-level end-to-end optimization of networked I/O for petascale distributed applications. Unlike existing approaches, it will treat data resources and the tasks related to data access and movement as first class entities just like computational resources and compute tasks, and not simply the side effect of computation.

Broader Impact:
This project will impact not just traditionally compute intensive disciplines from science and engineering, but also new emerging computational areas in the arts, humanities, business and education. The PI will be collaborating with other leading institutions in the area of distributed data management such as LBNL, ISI/USC, UNC, UCSD, and University Chicago/Argonne to integrate Stork with their data management solutions and disseminate it to their user communities. The comprehensive education component of the project will include science projects and summer training camps on data-intensive computing with K-12 students (where 99% are minority students), undergraduate and graduate student training, international student/intern exchange program, and minority workshops.

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5).

This project is to renovate the research component of Louisiana State University's (LSU's) campus network. The network includes a bypass feature which allows traffic between researchers and pre-cleared network addresses to bypass the institution's primary firewall and traffic-shaping devices. Such traffic flows between the border router and the campus network core through a router dedicated to bypass traffic, rather than through the campus firewall and traffic shaper. This architecture was created by LSU to address two primary concerns associated with the older network architecture, lack of adequate bandwidth over the 'last mile' to researchers with equipment generating or consuming large amounts of data in facilities in distributed locations, and the hindrance to high bandwidth, low-latency data flows imposed by various network security and integrity controls. The renovation involves upgrading multiple levels within the campus network. The goal is to upgrade the campus network in such a way that when, subsequently, a new high-bandwidth, low-latency connection is required at some location on campus, this can be deployed rapidly and inexpensively.

The upgraded network will enhance research in many areas, including coastal modeling, the visualization of coastal models, computational biology, relativistic astrophysics, computer science, advanced networking research, and nontraditional areas of computational study such as music, theatre, and the visual arts.

In addition to providing infrastructure for research, the upgraded network will also have an impact outside of science and engineering, on research related to technologies for use in the arts and digital media. The infrastructure will support research training since many of the people using the network in research activities will be graduate and undergraduate students.

This is an award to acquire a compute cluster at LSU. The computer is a heterogeneous HPC cluster named SuperMIC containing both Intel Xeon Phi and NVIDIA Kepler K20X GPU (graphics processing unit) accelerators. The intent is to conduct research on programming such clusters while advancing projects that are dependent on HPC. The efforts range from modeling conditions which threaten coastal environments and test mitigation techniques; to simulating the motions of tumors/organs in cancer patients due to respiratory actions to aid radiotherapy planning and management. The burden of learning highly complex hybrid programming models presents an enormous software development crisis and demands a better solution. SuperMIC will serve as the development platform to extend current programming frameworks, such as Cactus, by incorporating GPU and Xeon Phi methods. Such frameworks allow users to move seamlessly from serial to multi-core to distributed parallel platforms without changing their applications, and yet achieve high performance. The SuperMIC project will include training and education at all levels, from a Beowulf boot camp for high school students to more than 20 annual LSU workshops and computational sciences distance learning courses for students at LONI (Louisiana Optical Network Initiative) and LA-SiGMA (Louisiana Alliance for Simulation-Guided Materials Applications) member institutions. These include Southern University, Xavier University, and Grambling State University - all historically black colleges and universities (HBCU) which have large underrepresented minority enrollments. The SuperMIC cluster will be used in the LSU and LA-SiGMA REU and RET programs. It will impact the national HPC community through resources committed to the NSF XSEDE program and the Southeastern Universities Research Association SURAgrid. The SuperMIC will commit 40% of the usage of the machine to the XSEDE XRAC allocation committee.

Modern scientific research often involves the marshaling of very large data sets through extensive processing pipelines before usable information can be extracted. Examples of this type of data-intensive research include the near real-time analysis of time-varying medical images, decoding the genetic structure of diverse biological systems, simulating the merger of binary star systems, modeling complex flow structures in chemical reaction tanks, and determining how to effectively interact with extraordinarily high-resolution lunar and planetary data.

The CADIS project provides the critical networking infrastructure that enables this wide range of data-intensive research activities at Louisiana State University (LSU). CADIS links high-performance computer systems located on campus and at other state and national facilities to a new state-of-the-art building at LSU -- the Louisiana Digital Media Center (LDMC). Serving as a center for diverse, data-intensive research activities, the LDMC houses high resolution display technologies and custom visualization systems and is dedicated to digital computation, media production, and advanced teaching methods. CADIS enhances student learning experiences by facilitating the use of actual data sets from experimental instruments and large-scale computational simulations, and the classroom application of novel visualization methods. CADIS also is expected to open new opportunities for collaboration between LSU researchers and the growing digital media industry across Louisiana.

This CC-NIE Integration project performed at the Louisiana State University is developing a cyber-infrastructure integrating six different large scale scientific research groups with high performance computing (HPC) clusters at LSU using software defined network technology and Hadoop and MPI (Message Passing Interface) based parallel frameworks.

The project consists of three objectives: (i) Building 10Gbps software-defined network (SDN) with OpenFlow switches and controllers to provide multiple virtual network slices to each group; (ii) Transferring Big Data with automatically tuned operation through multiple optical paths over the SDN network achieving at most 20Gbps of aggregated disk-to-disk transfer rate; and (iii) Analyzing Big Data by developing data-intensive distributed computing frameworks with Hadoop and MPI technologies parallelizing large number of jobs over HPC clusters.

Those three components are integrated with a web portal service and a GENI-enabled network management system. To achieve high disk-to-disk transfer rate at 20Gbps, the project has an industrial partnership with Samsung Electronics that contributes 70TB Solid State Drive (SSD) storage and optimizes the I/O bandwidth.

The cyber-infrastructure accelerates the current Big Data research projects spanning a wide range of research areas including gene sequencing research at Biology and Vet School, computational chemistry, big data mining at Computer Science, coastal hazard simulation research at civil and environmental engineering. In the end, the project will establish a methodology to build integrated cyber-infrastructures consisting of high speed networks, high performance computing, and high speed storage for the Big Data Science and Engineering.