A5087049302- Scientific Computing and Data Management
- Distributed and Parallel Computing Systems
- Advanced Data Storage Technologies
- Parallel Computing and Optimization Techniques
- Cloud Computing and Resource Management
- Research Data Management Practices
- Quantum many-body systems
- Statistical Mechanics and Entropy
- Computational Physics and Python Applications
- Advanced Thermodynamics and Statistical Mechanics
- Mathematical and Theoretical Analysis
- Quantum Mechanics and Applications
- Model Reduction and Neural Networks
- ZnO doping and properties
- Electronic and Structural Properties of Oxides
- Data Quality and Management
- Black Holes and Theoretical Physics
- Data Visualization and Analytics
- Reservoir Engineering and Simulation Methods
- Noncommutative and Quantum Gravity Theories
- Electron and X-Ray Spectroscopy Techniques
- Particle Detector Development and Performance
- Distributed systems and fault tolerance
- Advanced MRI Techniques and Applications
- Computability, Logic, AI Algorithms
Argonne National Laboratory
2005-2025
University of Chicago
2007-2025
University of Illinois Chicago
2023-2025
University of California, Davis
2018
We present Swift, a system that combines novel scripting language called SwiftScript with powerful runtime based on CoG Karajan, Falkon, and Globus to allow for the concise specification, reliable efficient execution, of large loosely coupled computations. Swift adopts adapts ideas first explored in GriPhyN virtual data system, improving many regards. describe its use XDTM logical structure complex file structures. also services dispatch manage execution tasks parallel grid environments....
Process management is an extremely important concept in both business and scientific communities. Several workflow tools have been proposed recent years offering advanced functionality various domains. In the world, vendors offer commercial customized solutions targeting specific users. several open-source are freely available. However they directed toward service aggregation rather than distributed process management. Little consideration given to needs of client terms mapping flow client....
The act of sharing scientific knowledge is rapidly evolving away from traditional articles and presentations to the delivery executable objects that integrate data computational details (e.g., scripts workflows) upon which findings rely. This envisioned coupling process essential advancing science but faces technical institutional barriers. Whole Tale project aims address these barriers by connecting computational, data-intensive research efforts with larger process—transforming discovery...
Scripting accelerates and simplifies the composition of existing codes to form more powerful applications. Parallel scripting extends this technique allow for rapid development highly parallel applications that can run efficiently on platforms ranging from multicore workstations petascale supercomputers.
Abstract We review the efforts of Open Grid Computing Environments collaboration. By adopting a general three‐tiered architecture based on common standards for portlets and Web services, we can deliver numerous capabilities to science gateways from our diverse constituent efforts. In this paper, discuss support standards‐based using Velocity development environment. Our are abstraction layers provided by Java CoG kit, which hide differences different toolkits. Sophisticated services...
In this paper we present the Coaster System. It is an automatically-deployed node provisioning (Pilot Job) system for grids, clouds, and ad-hoc desktop-computer networks supporting file staging, on-demand opportunistic multi-node allocation, remote logging, monitoring. The System has been previously [32] shown to work at scales of thousands cores. used since 2009 applications in fields that include biochemistry, earth systems science, energy modeling, neuroscience. successfully on Open...
Exascale computers offer transformative capabilities to combine data-driven and learning-based approaches wi th traditional simulation applications accelerate scientific discovery insight. However, these software combinations integrations are difficult achieve due the challenges of coordinating deploying heterogeneous components on diverse massive platforms. We present ExaWorks project, which addresses many challenges. developed a workflow Software Development Toolkit (SDK), curated...
The practical realization of managing and executing large scale scientific computations efficiently reliably is quite challenging. Scientific often involve thousands or even millions tasks operating on quantities data, such data are diversely structured stored in heterogeneous physical formats, scientists must specify run over extended periods collections compute, storage network resources that heterogeneous, distributed may change constantly. We present the integration several advanced...
Exascale computers will offer transformative capabilities to combine data-driven and learning-based approaches with traditional simulation applications accelerate scientific discovery insight. These software combinations integrations, however, are difficult achieve due challenges of coordination deployment heterogeneous components on diverse massive platforms. We present the ExaWorks project, which can address many these challenges: is leading a co-design process create workflow Software...
Parallel scripting is a loosely-coupled programming model in which applications are composed of highly parallel scripts program invocations that process and exchange data via files. We characterize here the can benefit from on petascale-class machines, describe mechanisms make this feasible such systems, present results achieved with currently available petascale computers.
It is generally desirable for high-performance computing (HPC) applications to be portable between HPC systems, example make use of more performant hardware, effective allocations, and co-locate compute jobs with large datasets. Unfortunately, moving scientific systems challenging various reasons, most notably that have different schedulers. We introduce PSI/J, a job management abstraction API intended simplify the construction software components are over scheduler implementations. argue...
The Grid approach allows collaborative pooling of distributed resources across multiple domains. However, the benefits are limited to those offered by commodity application development framework used. Several elegant and flexible frameworks support only specific architectures, thereby not allowing applications exploit full potential Grid. In order initiate community interest standardize a high-level abstraction layer for different we introduce collection abstractions data structures that...
The Grid approach provides the ability to access and use distributed resources as part of virtual organizations. emerging infrastructure gives rise a class scientific applications services support collaborative resource-sharing requirements teleimmersion, visualization, simulation services. Because such operate in mode, data must be stored delivered timely manner meet deadlines. Hence, this has stringent real-time constraints quality-of-service (QoS) requirements. A QoS management is...
We have investigated the electronic structure of dilute magnetic semiconductor (DMS) $Ga_{0.98}Mn_{0.02}P$ and compared it to that an undoped $GaP$ reference sample, using hard X-ray photoelectron spectroscopy (HXPS) angle-resolved photoemission (HARPES) at energies about 3 keV. present experimental data, as well theoretical calculations, in order understand role Mn dopant emergence ferromagnetism this material. Both core-level spectra or angle-integrated valence are discussed. In...
The line between native and web applications is becoming increasingly blurred as modern browsers are powerful platforms on which can be run. Such trivial to install readily extensible easy use. In an educational setting, permit a way deploy tools in highly-restrictive computing environment.
We revisit the issue of geometrical separability Hilbert space physical states on lattice Abelian theories in context entanglement entropy. discuss conditions under which vectors space, as well gauge-invariant algebra, admit a tensor product decomposition with interpretation. With exception pure gauge lattices periodic boundary contain topological degrees freedom, we show that is geometrically separable.
This paper describes an ad hoc grid security infrastructure developed as a part of the Java CoG Kit project. It supports several requirements specific to sporadic nature grids. focuses on identity management, verification, and authorization control in spontaneous collaborations without pre-established policies or environments. adopts established community standards, with modifications where needed. also discusses integration implementation. The implementation secure collaboration grids using...