Estimation of parameters in compound Weibull distributions
Estimating procedure using electronic digital computers applicable to experiments in which mixed Weibull distribution is statistical model
SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Estimating procedure using electronic digital computers applicable to experiments in which mixed Weibull distribution is statistical model
In this article, sparse matrix vector multiply (SpMVM) is an important kernel that frequently arises in high performance computing applications. Due to its low arithmetic intensity, several approaches have been proposed in literature to improve its scalability and efficiency in large scale computations. In this paper, our target systems are high end multi-core architectures and we use messaging passing interface + open multiprocessing hybrid programming model for parallelism. We analyze the performance of recently proposed implementation of the distributed symmetric SpMVM, originally developed for large sparse symmetric matrices arising in ab initio nuclear structure calculations. We also study important features of this implementation and compare with previously reported implementations that do not exploit underlying symmetry. Our SpMVM implementations leverage the hybrid paradigm to efficiently overlap expensive communications with computations. Our main comparison criterion is the "CPU core hours" metric, which is the main measure of resource usage on supercomputers. We analyze the effects of topology-aware mapping heuristic using simplified network load model. Furthermore, we have tested the different SpMVM implementations on two large clusters with 3D Torus and Dragonfly topology. Our results show that the distributed SpMVM implementation that exploits matrix symmetry and hides communication yields the best value for the "CPU core hours" metric and significantly reduces data movement overheads.
The use of accelerators such as graphics processing units (GPUs) has become popular in scientific computing applications due to their low cost, impressive floating-point capabilities, high memory bandwidth, and low electrical power requirements. Hybrid high-performance computers, machines with nodes containing more than one type of floating-point processor (e.g. CPU and GPU), are now becoming more prevalent due to these advantages. In this paper, we present a continuation of previous work implementing algorithms for using accelerators into the LAMMPS molecular dynamics software for distributed memory parallel hybrid machines. In our previous work, we focused on acceleration for short-range models with an approach intended to harness the processing power of both the accelerator and (multi-core) CPUs. To augment the existing implementations, we present an efficient implementation of long-range electrostatic force calculation for molecular dynamics. Specifically, we present an implementation of the particle-particle particle-mesh method based on the work by Harvey and De Fabritiis. We present benchmark results on the Keeneland InfiniBand GPU cluster. We provide a performance comparison of the same kernels compiled with both CUDA and OpenCL. We discuss limitations to parallel efficiency and future directions for improving performance on hybrid or heterogeneous computers.
This report summarizes research conducted at the Institute for Computer Applications in Science and Engineering in the areas of (1) applied and numerical mathematics, including numerical analysis and algorithm development; (2) theoretical and computational research in fluid mechanics in selected areas of interest, including acoustics and combustion; (3) experimental research in transition and turbulence and aerodynamics involving Langley facilities and scientists; and (4) computer science.
This paper describes new and recent advances in the development of a hybrid transfinite element computational methodology for applicability to conduction/convection/radiation heat transfer problems. The transfinite element methodology, while retaining the modeling versatility of contemporary finite element formulations, is based on application of transform techniques in conjunction with classical Galerkin schemes and is a hybrid approach. The purpose of this paper is to provide a viable hybrid computational methodology for applicability to general transient thermal analysis. Highlights and features of the methodology are described and developed via generalized formulations and applications to several test problems. The proposed transfinite element methodology successfully provides a viable computational approach and numerical test problems validate the proposed developments for conduction/convection/radiation thermal analysis.
The use of coprocessors or accelerators such as graphics processing units (GPUs) has become popular in scientific computing applications due to their low cost, impressive floating-point capabilities, high memory bandwidth, and low electrical power re- quirements. Hybrid high-performance computers, defined as machines with nodes containing more than one type of floating-point processor (e.g. CPU and GPU), are now becoming more prevalent due to these advantages. Although there has been extensive research into methods to efficiently use accelerators to improve the performance of molecular dynamics (MD) employing pairwise potential energy models, little is reported in the literature for models that include many-body effects. 3-body terms are required for many popular potentials such as MEAM, Tersoff, REBO, AIREBO, Stillinger-Weber, Bond-Order Potentials, and others. Because the per-atom simulation times are much higher for models incorporating 3-body terms, there is a clear need for efficient algo- rithms usable on hybrid high performance computers. Here, we report a shared-memory force-decomposition for 3-body potentials that avoids memory conflicts to allow for a deterministic code with substantial performance improvements on hybrid machines. We describe modifications necessary for use in distributed memory MD codes and show results for the simulation of water with Stillinger-Weber on the hybrid Titan supercomputer. We compare performance of the 3-body model to the SPC/E water model when using accelerators. Finally, we demonstrate that our approach can attain a speedup of 5.1 with acceleration on Titan for production simulations to study water droplet freezing on a surface.
In a nonlocal game, two noncommunicating players cooperate to convince a referee that they possess a strategy that does not violate the rules of the game. Quantum strategies allow players to optimally win some games by performing joint measurements on a shared entangled state, but computing these strategies can be challenging. We present a variational quantum algorithm to compute quantum strategies for nonlocal games by encoding the rules of a nonlocal game into a Hamiltonian. We show how this algorithm can generate a short-depth optimal quantum strategy for a graph coloring game with a quantum advantage. This quantum strategy is then evaluated on fourteen different quantum hardware platforms to demonstrate its utility as a benchmark. Finally, we discuss potential sources of errors that can explain the observed decreased performance of the executed task and derive an expression for the number of samples required to accurately estimate the win rate in the presence of noise.
The efficient utilization of mixed-precision numerical linear algebra algorithms can offer attractive acceleration to scientific computing applications. Especially with the hardware integration of low-precision special-function units designed for machine learning applications, the traditional numerical algorithms community urgently needs to reconsider the floating point formats used in the distinct operations to efficiently leverage the available compute power. In this study, we provide a comprehensive survey of mixed-precision numerical linear algebra routines, including the underlying concepts, theoretical background, and experimental results for both dense and sparse linear algebra problems.
Computer application to flat earth problem with analytical functions - guidance systems
A nonperturbative analytic solution of the high charge and energy (HZE) Green's function is used to implement a computer code for laboratory ion beam transport. The code is established to operate on the Langley Research Center nuclear fragmentation model used in engineering applications. Computational procedures are established to generate linear energy transfer (LET) distributions for a specified ion beam and target for comparison with experimental measurements. The code is highly efficient and compares well with the perturbation approximations.
Multi-dimensional MPI communications, where MPI communications have to be performed in each dimension of a Cartesian communicator, have been frequently used in many of today's high performance computing applications. While individual MPI collective communications for regular communicators with a one-dimensional linear-ranking of processes have been extensively studied and optimized, little optimizations have been performed for multi-dimensional MPI collective communications on multi-dimensional Cartesian topology. In this paper, we optimize multi-dimensional MPI collective communications for SMP and multi-core systems at the application level. We show that the default Cartesian topology built by the state-of-the-art MPI implementations produce sub-optimal performance for multi-dimensional MPI collective communications. We design optimal process-to-core mapping schemes for Cartesian communicators to minimize the total inter-node communications. The proposed technique improves the performance by up to 80% over the default Cartesian topology built by Cray's MPI implementation MPT 3.1.02 on the world's current second fastest supercomputer Jaguar at Oak Ridge National Laboratory.
ZFP is a floating-point compression format gaining traction in high-performance computing applications. A software implementation has demonstrated the ability to reduce data movement across communication channels and to reduce the footprint of floating-point arrays in memory. Nevertheless, any benefit in performance is limited to the spare compute cycles available before reaching bandwidth limitations. A hardware implementation of ZFP has the potential to substantially improve HPC application performance. Hardware compression speedup over software alone on a single CPU core ranges from 15x for 1-D arrays to over 200x for 3-D arrays. The implementation of ZFP is sourced in SystemC to facilitate its evaluation in various architectures. The encode pipeline consists of several modules. Uncompressed blocks of floating-point numbers in IEEE format flow into the pipeline in a stream and a compressed bitstream flows out from the pipeline. The decode pipeline is the inverse of the encode pipeline. A modular design enables formats other than IEEE (such as posits) to be considered with minor adaptations. Best performance will be realized from the hardware ZFP unit when batches of blocks are processed at a time. The implementation supports 1-D, 2- D and 3-D blocks of floating-point numbers. The hardware implementation of ZFP has been validated with the software implementation of ZFP. C++ template parameters are used to specify the bit width of floating-point numbers and the array dimension of the encoder. A test bench program has been created with several test cases, some with continuous data that is ideal for ZFP, and others with extreme cases containing numbers near or at the maximum or minimum values supported by the number format.
Recent advances in nanotechnology have enabled researchers to manipulate small collections of quantum-mechanical objects with unprecedented accuracy. In semiconductor quantum-dot qubits, this manipulation requires controlling the dot orbital energies, the tunnel couplings, and the electron occupations. These properties all depend on the voltages placed on the metallic electrodes that define the device, the positions of which are fixed once the device is fabricated. While there has been much success with small numbers of dots, as the number of dots grows, it will be increasingly useful to control these systems with as few electrode voltage changes as possible. Here, we introduce a protocol, which we call the "compressed optimization of device architectures" (CODA), in order both to efficiently identify sparse sets of voltage changes that control quantum systems and to introduce a metric that can be used to compare device designs. As an example of the former, we apply this method to simulated devices with up to 100 quantum dots and show that CODA automatically tunes devices more efficiently than other common nonlinear optimizers. To demonstrate the latter, we determine the optimal lateral scale for a triple quantum dot, yielding a simulated device that can be tuned with small voltage changes on a limited number of electrodes.
A nonperturbative analytic solution of the high charge and energy (HZE) Green's function is used to implement a computer code for laboratory ion beam transport in multilayered materials. The code is established to operate on the Langley nuclear fragmentation model used in engineering applications. Computational procedures are established to generate linear energy transfer (LET) distributions for a specified ion beam and target for comparison with experimental measurements. The code was found to be highly efficient and compared well with the perturbation approximation.
Cloud object storage is enabled for archived data, such as checkpoints and results, of high performance computing applications using a middleware process. A plurality of archived files, such as checkpoint files and results, generated by a plurality of processes in a parallel computing system are stored by obtaining the plurality of archived files from the parallel computing system; converting the plurality of archived files to objects using a log structured file system middleware process; and providing the objects for storage in a cloud object storage system. The plurality of processes may run, for example, on a plurality of compute nodes. The log structured file system middleware process may be embodied, for example, as a Parallel Log-Structured File System (PLFS). The log structured file system middleware process optionally executes on a burst buffer node.
High performance computing applications are producing increasingly large amounts of data and placing enormous stress on current capabilities for traditional post-hoc visualization techniques. Because of the growing compute and I/O imbalance, data reductions, including in situ visualization, are required. These reduced data are used for analysis and visualization in a variety of different ways. Many of he visualization and analysis requirements are known a priori, but when they are not, scientists are dependent on the reduced data to accurately represent the simulation in post hoc analysis. The contributions of this paper is a description of the directions we are pursuing to assist a large scale fusion simulation code succeed on the next generation of supercomputers. Finally, these directions include the role of in situ processing for performing data reductions, as well as the tradeoffs between data size and data integrity within the context of complex operations in a typical scientific workflow.
The Office of Management and Budget (OMB) Circular A-71, transmittal Memorandum No. 1, requires that each agency establish a management control process to assure that appropriate administrative, physical and technical safeguards are incorporated into all new computer applications. In addition to security specifications, the management control process should assure that the safeguards are adequate for the application. The security activities that should be integral to the system development process are examined. The software quality assurance process to assure that adequate and appropriate controls are incorporated into sensitive applications is also examined. Security for software packages is also discussed.
This report covers the activities carried out by the project The ACTS Center for Software Sustainability, whose proposal was submitted to ASCR in 2007. These activities have their roots in the project Advanced Computational Testing Software (ACTS) Toolkit, which was part of the DOE2000 Program. Driven by the state of the software, the technology and the nature of the tools in the Toolkit, the project was reformulated in a project that better represented the goals and activities within ACTS, An Expanded Framework for the Advanced CompuTational Software (ACTS) Collection. At the heart of this project was the consolidation of lessons learned and demands from the computational science community that guided us to the creation of the now well-known DOE ACTS Collection. The goal of the project was to increase the pay-off of DOE ASCR investment in the development of the general purpose tools in the DOE ACTS Collection, reduce the duplication of efforts, and minimize the development time of high-end computer applications while maximizing the life of the codes.