 |
 |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Current Scientific Research ProjectsOur projects are grant-funded research and development efforts, designed to answer specific research questions or provide tools and techniques to the scientific community. Tech-X Corporation collaborates with the U.S. Department of Energy, renowned independent research labs like Brookhaven National Laboratory and institutions such as the University of Colorado, producing significant solutions to real-world problems. For additional information, please see the Web site of the appropriate funding agency. Online CollaborationWe have a separate Web site designed to allow easy collaboration between Tech-X and outside scientists and developers. This site is located at https://collaborate.txcorp.com. If you would like to contribute to this site, please contact the PI of the project in which you are interested. Project ListingsFusion Services Group
Distributed Technologies Group Accelerator Technology Group
Beam-Plasma Interactions Group Space Applications Group
BPPS: Building Blocks for the Rapid Development of Parallel Simulation Scientists need to be able to quickly develop and run parallel simulations without paying the high price of writing low-level message passing codes using compiled languages such as C/C++/Fortran. This is especially true of students and researchers who are expert scientists in their field but don't have the time or resources to become experts in parallel computing. This project will enable scientists to rapidly create parallel simulations by providing parallel building blocks in the high-level programming language Python. The building blocks themselves, such as distributed arrays, parallel linear algebra, parallel Fourier transforms and parallel statistical algorithms, will be implemented by leveraging existing high-performance libraries and creating high-level Python objects that completely hide the details of the underlying libraries and parallelism from users. The building blocks will be developed as part of the open source IPython project, which will enable users to develop, debug and run parallel simulations in a completely interactive manner, similar to Matlab or IDL.
COMPASS SciDAC-2 Project Tech-X Corporation is a key member of the multi-institution Community Petascale Project for Accelerator Science and Simulation (COMPASS). Tech-X personnel are working in three technical areas to help address priorities of the DOE Office of Science, Offices of High Energy Physics (HEP) and Nuclear Physics (NP): 1) electromagnetic simulation and design of complex superconducting accelerator structures with intense relativistic particle beams; 2) simulation of present and design of future experiments for laser-plasma acceleration of electrons to high energy in short distances; and 3) simulating the microphysics of standard and "coherent" electron cooling systems for increasing the luminosity of relativistic heavy-ion colliders. All of these efforts depend on the parallel VORPAL framework, now routinely using >1,000 processors at NERSC and other leadership computing facilities, with the goal of simultaneously and efficiently using ~100,000 processors by 2012. The principal investigator for this project is John Cary.  Back to
Top
User Centric Monitoring Software services and components will be developed that provide a rich set of Web-accessible intuitive information to scientists. Benefits include greater productivity for the scientists, by allowing them to monitor computing tasks distributed throughout the Grid. Back to
Top
SuRF:Three-Dimensional Self-Consistant Simulations of Multipacting in Superconducting Radio Frequencies Improving performance of superconducting radio frequency accelerating cavities will decrease the cost of next-generation linear particle accelerators. Innovative software will be developed to improve the modeling and design of superconducting radio frequency cavities. Back to
Top
ISENS: Information Processing System for Sensor Data Intelligence analysts are required to process ever growing amounts of sensor related data for situation assessment and decision making. In this project an innovative new software system will be developed to improve the efficiency of processing and understanding of such data. Back to
Top
High Z Droplets: A Novel Source of Heavy Ions for Nuclear Physics High-fidelity software will be developed for dramatic improvement of rare-isotope ion source technology, required for fundamental advances in nuclear physics and astrophysics research. Back to
Top
GSIMF: A Grid Software Installation Management Framework This project will develop tools for remote, secure management of software programs on a large computer network. These tools should increase the efficiency with which scientists work and share resources. Back to
Top
Fusion Framework: Framework for Modernization and Componentization of Fusion Modules This project will develop software tools that will increase software reuse and data interoperability as well as facilitate integrated scientific modeling, collaborative data analysis, and archiving of fusion data. These tools will help to ensure successful participation of the US in the ITER project. Back to
Top
ROVER: Remote Online Visualization Environment for Researchers In modern research environments it is common for researchers to want to visualize remote data sets that are large enough to prohibit transfer to local resources for rendering. This project will develop an online environment that allows for this type of visualization requiring the researcher to only have access to a Web browser. The online environment will also allow researchers distributed geographically to collaboratively visualize remote data. Back to
Top
RF Synergia: Modeling Accelerator Beam Dynamics Including Superconducting RF Cavities Through computer modeling this project will enable researchers to ensure that superconducting rf cavities in modern particle accelerators will work as they are designed to work. Back to
Top
Synergia: A Beam Dynamics Application Based on the Common Component Architecture Particle accelerators are used as fundamental tools in scientific investigations that encompass the entire purview of the DOE s Office of Science. The DOE has made a significant financial investment in the development of software tools and libraries for the purpose of simulating and modeling current and next-generation accelerators. These software resources could be used in a wider range of accelerator modeling software, if they were made available in the form of CCA components. This CCA/Synergia project is developing a CCA-compliant, component-based beam dynamics application. The components that comprise this application are being created from a set of services that are currently provided by two separate science packages (Synergia and MaryLie/Impact) and one high-performance computer science package (PETSc). Back to
Top
Diamond Amp: Simulation Package for Parallel 3D Modeling of an Electron Gun with a Diamond Amplifier Novel high-current electron sources are required for major upgrades of existing particle accelerators, in order to further advance the field of experimental nuclear physics. High-fidelity software is being developed to enable more accurate design of a recent concept for such sources. Back to
Top
Accurate Numerical Models of the Secondary Electron yield from Grazing-Incidence Collisions Interactions with unwanted electrons are a major limiting factor in the performance of ion accelerators, and collisions at grazing incidence between ions and beam pipe walls are a main source of these unwanted electrons. Computer modeling could be used to examine this problem, but the codes in the heavy-ion fusion community presently do not have the capability to accurately model grazing-incidence collisions. In this project, Tech-X Corporation is developing accurate numerical models of grazing-incidence collisions between ions and walls. These models will give heavy-ion fusion simulation codes the ability to determine ways to mitigate the effects of unwanted electrons. Phase I will focus on the development of prototype numerical models using state-of-the-art simulation techniques, with model validation against recent heavy-ion fusion accelerator simulations in order to evaluate performance effects of grazing-incidence ion-wall collisions. Learn about TxPhysics Library, a cross-platform library of computational modules for studying electron effects in accelerators. Back to Top
RHIC: Magnetized Electron Transport in the Proposed Electron Cooling Section of the RHIC With a grant from the DOE, Tech-X Corporation is working with Brookhaven National Laboratory to address the need for their electron cooling system (currently under research and development) to perform high-speed, high-fidelity numerical simulations that assess the impact of space-charge, nonlinearities, and machine errors as part of the research and design process. Tech-X Corporation is adding needed functionality to an existing parallel 3-D code, making it a suitable tool for doing realistic simulations and analysis of electron cooling systems for relativistic ions. The project focuses on demonstrating the ability to handle misalignments and RF cavity nonlinearities correctly for the purposes of tracking simulations, and showing the usefulness of the added functionality by applying it to the electron cooling system. Back to Top
A Data Skimming Grid Portal High-energy physics data sets are currently very large and will continue to grow. A time consuming and labor-intensive stage of any experimental effort is the event selection (or skimming) that must be performed on a remote site in order to deliver reasonably sized data sets to the end-user. Tech-X Corporation is addressing this need by developing a Web-invocable Grid Service tool called TxFlow for configuring executions of the CMS-SW software framework used for data analysis purposes such as skimming. This TxFlow tool is written in Java and allows scientist to visually configure the CMS-SW by selecting icons for a tree and dragging them to a canvas to create workflows. Currently in Phase II, this project underway in improving the prototype application to work specifically with the CMS-SW framework while also being generically configurable to work with many workflows through the use of XML. More information about the project Back to Top
CMEE: Computational Modules for Studying Electron Effects in Heavy Ion Fusion Accelerators The electron cloud effect is a major limiting factor in performance of proton accelerators. To ensure this does not adversely affect the performance of a heavy-ion fusion accelerator, studying electron effects in the regimes relevant to heavy-ion fusion is essential prior to building a reliable proton accelerator. Computer modeling is the most widely used method of studying the problem, but the main codes used in the heavy-ion fusion community presently are not capable of studying the electron cloud effect. Utilizing a DOE grant, Tech-X Corporation is developing a cross-platform library of cross-platform, cross-language numerical libraries for modeling particle interactions with plasmas and solids with applications to beam dynamics and high energy density physics. This library offers heavy-ion fusion simulation codes access to the most recent experimental data and numerical routines for modeling the electron cloud effect. The library will have a concise, canonical interface along with documentation, and will be portable to all major platforms and all major heavy-ion fusion research codes. Learn about TxPhysics Library, a cross-platform library of computational modules for studying electron effects in accelerators. SPECS: Smart Particle Electron Cooling Simulations The Relativistic Heavy Ion Collider (RHIC) at Brookhaven National Laboratory collides heavy ions to create conditions similar to those that existed a fraction of a second after the big bang. To improve the performance of this premier nuclear physics facility, DOE scientists are building an electron cooling section as part of a planned luminosity upgrade. This electron cooling section is fundamentally different than those built at other accelerator facilities, so DOE scientists need the capability to conduct high-performance, high-fidelity numerical simulations during the research and design process. To help DOE scientists conduct these detailed simulations, Tech-X Corporation is developing a parallel 3-D particle simulation solution that offers novel features such as a "smart" particle algorithm. This algorithm incorporates the detailed microphysics of magnetized Coulomb collisions with minimal overhead, enabling the simultaneous capture of space charge effects and thermal energy transfer - even for complicated electron and ion distributions. Learn more about VORPAL. Back to Top
GRETINA: High Performance Algorithms for Signal Decomposition in Gamma Ray Detectors Gamma ray detectors play an important role in a broad range of science and technology applications, including astrophysics and medical imaging. To make these applications more accurate and efficient, scientists are building GRETA (Gamma Ray Energy Tracking Array) - a segmented gamma ray detector. An important stepping stone to GRETA is the GRETINA gamma ray detector array, which consists of Ge crystal diodes that produce currents upon impact of gamma rays. By decomposing these currents into combinations of known signals, scientists can determine the number, location, and energies of the gamma ray interactions. The challenge is that decomposition must be done in real-time as the experiment is running, which requires fast and efficient algorithms. While GRETINA promises to be a valuable source of new science for the nuclear physics community, solving the signal decomposition challenge is vital to the project's success. To address this challenge, Tech-X Corporation is identifying the data structures and algorithms that are best suited to fast and efficient signal decomposition of GRETINA data and then optimizing the algorithms for speed on specific hardware platforms. FastCTR (Fast Coherent Transition Radiation): Rapid 3D Simulation of a Bunch-Length Diagnostic for Laser Wakefield Accelerators via Coherent Transition Radiation at THz Frequencies Laser wakefield accelerator (LWFA) concepts - characterized by extremely short particle beams - show great promise for reducing the cost and size of future-generation high-energy linear accelerators. However, to continue rapidly advancing LWFA technology, scientists need access to a non-invasive bunch-length diagnostic. The coherent transition radiation (CTR) generated as short particle beams exit the plasma provides such a diagnostic. In this project, Tech-X Corporation is enhancing existing particle-in-cell (PIC) simulation software to enable the measurement of ultra-short electron beams and conducting PIC simulations using 2- and 3-D Cartesian geometry to characterize the CTR emitted from a self-modulated (SM) LWFA. The enhanced PIC simulation code, coupled with the experimental measurements, delivers a uniquely powerful diagnostic for plasma-based accelerators. Back to Top
FTRTNS: A Fault-Tolerant Real-Time CORBA Naming Service Many mission critical applications are required to meet high degrees of dependability and predictability requirements simultaneously. It is still not possible to use CORBA to manage applications' FT and RT properties in unison due to contradicting strategies in the support mechanisms. The emerging OMG's Lightweight Fault-Tolerance for Distributed Real-Time Systems specification aims to address this challenge by defining standard interactions between CORBA and external FT mechanisms while maintaining the portability of applications based on CORBA middleware. This project will investigate key technical issues and implement a prototype Lightweight Fault Tolerance CORBA based on the emerging specification, and develop example FT and RT applications using the prototype. Back to Top
Distributed Components: Distributed CCA Components and Grid Services for Scientific Computing Scientists working on the Fusion Simulation Project want to be able to reliably and comprehensively predict fusion plasma dynamics and properties by integrating legacy codes that, individually, adequately describe isolated problems. However, combining codes to build higher level, more comprehensive models is currently impractical. Tech-X Corporation is enhancing existing implementations of the Common Component Architecture (CCA) framework to create new software tools that facilitate large-scale computer simulations of the physical processes in a thermonuclear reactor. As part of this project, Tech-X researchers will demonstrate that a Remoting Component can be used to enable distributed applications using existing frameworks. Tech-X Corporation will show that the Native Array library facilitates the conversion of legacy code into components. Back to Top
FSML: Fusion Simulation Mark-up Language for Interoperability of Data and Analysis Tools One goal of the Fusion Simulation Project is to create a flexible framework that combines data generated by multiple simulations in the fusion and plasma physics community. These simulations must be able to seamlessly exchange data with one another. However, interoperability is a challenge because the data formats and data analysis tools used in the fusion and plasma physics simulations are highly heterogeneous. To address this problem, Tech-X Corporation is developing the Fusion Simulation Markup Language (FSML). Based on the eXtensible Markup Language (XML), FSML describes and accesses simulation data of various formats to facilitate integrated scientific modeling, collaborative data analysis and archiving of fusion data. This scalable solution enables comparison of results from different simulations, resulting in a deeper understanding of physics. More information about the project Back to Top
DSMC-PIC: A Parallel PIC-DSMC Code for Modeling Complex Plasmas Plasma propulsion systems for satellites, the low-density plasma edge in fusion devices, and plasma reactors for semiconductor manufacture are all in need of a modeling and simulation capability that effectively addresses electromagnetics, charged particle dynamics, neutral gas dynamics, and the interaction among these entities. Moreover, such simulation capability should be able to handle fully three-dimensional geometries with complex boundaries. To fill this need, Tech-X Corporation is utilizing a grant from the DOE to augment existing software that models arbitrary dimensional plasmas but lacks the capability to model background gases or complex boundaries. Using simple algorithms, the project includes a Direct-Simulation Monte-Carlo (DSMC) capability that allows modeling background gases at high Knudsen number, and a Navier Stokes capability that models background gases at low Knudsen number (the fluid limit). Learn more about VORPAL. Back to Top
UCM: Grid Service for User-Centric Job Monitoring Nuclear and high-energy physicists routinely execute data processing and data analysis jobs on a Grid and need to be able to monitor their jobs execution at an arbitrary site at any time. Existing Grid monitoring tools provide abundant information about the whole system, but are geared towards production jobs and well suited for Grid administrators, while the information tailored towards an individual user is not readily available in a user-friendly and user-centric way. Tech-X proposes to develop a framework composed of a library, database services, and a Web portal that will collect and filter available job monitoring information from various resources and present it to users in a user-centric view. Virtual organizations can use the proposed framework to provide physicists with the ability to monitor such information as the status of the submitted job, queue position, time of the start/finish, percentage of being done, error messages, standard output, and reasons for failure. A working prototype for the Grid service for User-Centric Job Monitoring was created. We determined the basic list of job properties that user would be interested in monitoring. We created an interface for the Grid service to expose the monitoring information and built a prototype service. We also built a prototype Web portal to display the information to users after they are varied from their Grid credentials. In Phase II, we will build a fully functional User-Centric Monitoring (UCM) Framework that will allow scientists to track jobs in a workflow through a Web portal with more information than is currently available such as reliable fine-grained job status, sometime hidden output and error messages from applications, scheduler queue position, and estimated job and task progress. The UCM Framework will provide a flexible library, services, and portlets to integrate within a Virtual Organization's existing Grid systems. Monitoring properties will be stored in a database for provenance and presented through the user's Web browser. Back to Top
FEL SRF: Beam Breakup Modeling for Energy Recovering Linacs BBU (Beam Breakup Instability) occurs when a normal accelerating and energy recovering cavity hosts an undesirable cavity mode, which disrupts the beam via induced transverse motion. In turn, the beam's perturbed motion may feed energy back into that same mode, causing an instability, potential loss of beam, and undesirable thermal loading on the accelerator structures. This project is developing new multi-mode/multi-cavity/multi-pass beam breakup threshold analysis capability, based upon full-geometry particle-and-fields time-domain simulation of the cavity structures. Accurate cut-cell modeling of the cavities, combined with robust PIC treatment of intense relativistic electron beams, is required because of the potential for many high order modes (HOM) to be present simultaneously, and the possibility for a variety of different modes to dominate. Back to Top
SciDAC RF: Time-Domain Particle-in-Cell Simulations of the Edge Plasma Region in the Vicinity of the RF Antenna and Nearby Structures Modeling of Ion Cyclotron and Lower Hybrid heating and current drive in tokamaks is usually done in the frequency domain, at the single frequency excitation of the heating source. This leads to a linear analysis of the wave propagation problem at the fast rf scale. However, it is known that several non-linear effects can occur, especially in the edge regions of the plasma, where RF sheaths and PDI (parametric decay instability) can anomalously divert the heating power away from the core. In this project Tech-X is developing time-domain methods that can treat the complex wave propagation scenarios, including resonances, cutoffs, and mode-conversions, while at the same time can incorporate these non-linear effects. These non-linear edge effects are expected to be strongly influenced by 3-D non-axisymmetric geometry of the edge region and coupling structures. Hence, the time-domain plasma model is being incorporated into the Vorpal Simulation framework, to take advantage of its advanced cut-cell geometry representations. Back to Top
MMM Adding individual solid particles to a dense array of polymer coils creates complex and poorly understood packing problems that span nanometer to micron length scales. Establishing a theoretical connection between polymer radius of gyration, particle size, shape, and dispersibility, and correlating these factors with mechanical response and other properties represents a tremendous fundamental challenge. Further complications and opportunities are afforded by introducing molecular scale segregation associated with block and graft copolymers. Scott Sides will work with the Polymer Chemistry Group to help address these issues by using modern computational methods based on self consistent field theoretic (SCFT) approaches and/or hybrid methods to examine model particles and polymers. Back to Top
New BC: New Boundry Algorithms for Next Generation Simulations and Design of High-Power Microwave Devices The goal of the project is to identify, implement and test algorithms for handling complex geometries within electromagnetic particle-in-cell codes (EM-PIC). This is motivated by the need to model high-power microwave (HPM) devices such as magnetrons. Modern magnetrons involve complex geometries which require the computational capacity to handle the associated boundaries. Most EM-PIC codes use a uniform Cartesian mesh which presents challenges to modeling complex structures for both the electromagnetic models and the particle-in-cell models. Due to the constant interactions of the electrons with the walls in these HPM devices simple stair step boundaries are not sufficient to understand the electron and field behavior near the boundaries. Back to Top
TASCS: Technology for Advanced Scientific Component Software The initial SciDAC initiative developed the Common Component Architecture (CCA) and brought the benefits of component-based software engineering to high-performance scientific software. Scientific teams who have adopted the CCA are now realizing the advantages of this extensible environment, which facilitates software interoperability within and across scientific domains, addressing issues in programming language interoperability, domain-specific common interfaces, and dynamic composability. Teams increasingly report that the CCA has become integral to the future of their science. We propose to extend the software component methodology, in close collaboration with a number of key application projects, through an interlinked series of initiatives, leveraging the component environment to develop powerful new capabilities. The initiatives focus on coupling parallel simulations, supporting emerging hardware and software paradigms for petascale computing, enhancing software quality and robustness, and dynamically adapting applications. We will continue to enhance the core CCA software environment, with emphasis on improving usability, and we will build a component ecosystem to provide more off-the-shelf components. Outreach activities include tutorials and other educational activities as well as collaborations with numerous applications, Centers for Enabling Technology, and Institutes. Back to Top
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© Tech-X Corporation 2004-2008 |
