Search NASA⌕ Search

SEARCH · Search NASA

Results for “Distributed memory”

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.

At least 559 records · Page 31

Center for Programming Models for Scalable Parallel Computing

Rice University's achievements as part of the Center for Programming Models for Scalable Parallel Computing include: (1) design and implemention of cafc, the first multi-platform CAF compiler for distributed and shared-memory machines, (2) performance studies of the efficiency of programs written using the CAF and UPC programming models, (3) a novel technique to analyze explicitly-parallel SPMD programs that facilitates optimization, (4) design, implementation, and evaluation of new language features for CAF, including communication topologies, multi-version variables, and distributed multithreading to simplify development of high-performance codes in CAF, and (5) a synchronization strength reduction transformation for automatically replacing barrier-based synchronization with more efficient point-to-point synchronization. The prototype Co-array Fortran compiler cafc developed in this project is available as open source software from http://www.hipersoft.rice.edu/caf.

97 MATHEMATICS AND COMPUTING↗

Using the Global Arrays Toolkit to Reimplement NumPy for Distributed Computation

Global Arrays (GA) is a software system from Pacific Northwest National Laboratory that enables an efficient, portable, and parallel shared memory programming interface to manipulate distributed defense arrays. Using a combination of GA and NumPy, we have reimplemented NumPy as a distributed drop-in replacement called Global Arrays in NumPy (GAiN). Scalability studies will be presented showing the utility of developing serial NumPy codes which can later run on more capable clusters or supercomputers.

Global Arrays↗

Investigation of chemical distribution in the oxide bulk layer in Ti/HfO{sub 2}/Pt memory devices using x-ray photoelectron spectroscopy

Resistive switching (RS) of Ti/HfO{sub 2}/Pt memory devices was studied using X-ray photoelectron spectroscopy. Hf{sup 4+} monotonously decreases with depth increasing towards to HfO{sub 2}/Pt interface in low resistance state, while a fluctuation distribution of Hf{sup 4+} is shown in high resistance state (HRS) and in the pristine Ti/HfO{sub 2}/Pt devices (without any SET or RESET process). It is explained by the existence of locally accumulated oxygen vacancies (clusters) in the oxide bulk layer in HRS and pristine states. A dynamic model of RS processes was proposed that the oxygen vacancy clusters dominantly determines the resistivity by the connecting/rupture between the neighbor cluster sites in the bulk.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

On-the-fly Doppler Broadening with Probability Table Interpolation for Unresolved Resonance Region in RMC

With the rapid development of computational power and parallel algorithms, Monte Carlo method has been widely investigated and used in neutron transport simulations due to its advantages in geometry modelling and usage of continuous energy point-wise cross sections. Nowadays the road has been paved to couple Monte Carlo codes with thermal-hydraulics codes to make multi-physics calculations. One of the most important aspects of thermal-hydraulics feedback is the detailed temperature distribution, which demands temperature dependent cross sections in neutron transport simulations. The conventional pre-generated cross section libraries at each temperature point are no longer feasible for detailed temperature distributions because of tremendous memory footprint. Hence, different on-the-fly Doppler broadening methods have been proposed for resolved resonance energy region and thermal energy region, such as the interpolation, regress model fitting, multipole representation (MPR), target motion sampling (TMS) and different integration algorithms based on Doppler broadening equation. Compared to resolved resonance energy region and thermal energy region, the representation of cross sections in unresolved resonance region is slightly different. As a result of the limitation of experiment measurements, the overlapped resonances cannot be distinguished from each other and the statistic average resonance parameters are given in the evaluated nuclear data file. In early applications, infinite-dilute pointwise cross sections were obtained from average parameters and were directly used in neutron simulations. However, it was proven that these infinite-dilute pointwise cross sections cannot capture the energy self-shielding effect, thus leading to an overestimate of reaction rates and usually an underestimate of k{sub eff} owing to the increased absorption rates and increased scattering probability to energy below unresolved resonance region where the capture-to-fission ratio is typically higher. This phenomenon is of great importance especially for problems where there is a considerable flux in unresolved resonance region. Then the probability table method was put forward by Levitt to take energy self-shielding effect into account, which provides probability distribution function of cross sections in unresolved resonance region instead of specific cross section values. The probability distribution function is presented in the form of tables and the mean values are equal to the infinite dilution cross sections. To obtain the temperature dependent cross sections in unresolved energy region, there are two different methods. One is to calculate cross sections directly from average resonance parameters and their statistical distributions , which is effective and precise but time-consuming. The other is to interpolate between pre-generated probability table files, which is much more efficient. In this paper, on-the-fly Doppler broadening of cross sections in unresolved resonance regions based on probability tables interpolation was implemented in RMC codes as a supplement to on-the-fly Doppler broadening for resolved resonance energy region and thermal region, which is based on the integration of Doppler broadening equation. The results of criticality calculation and depletion calculation show that on-the-fly Doppler broadening for the whole energy region is accurate and efficient. (authors)

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Qutrits in multiparticle systems

The evolution of complex composite quantum systems which can be reduced to systems with a small dimensionality of the Hilbert space (of the qutrit, ququart type, etc.) is considered. In the case of the interaction of an ensemble of two-level atoms with light, the conditions are found under which a qutrit is produced from the light and atomic states. The properties and possible applications of a qutrit based of the Fock states of light in which two photons are distributed among three modes are discussed. It is shown that this state has the nonclassical photon statistics, is entangled and can be used as a quantum channel for the teleportation, dense coding, and key distribution. (fifth seminar in memory of d.n. klyshko)

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Exploring Architectural-Aware Affinity Policies in Modern HPC Runtimes

Modern commodity and High-Performance Computing (HPC) systems are evolving with complex CPU architectures. These architectures now feature higher core and NUMA domain counts and implement features such as hyperthreading. When considering significant differences in hardware configurations, library availability, and hardware-tailored system/software stacks, which could substantially vary from one system to another, performance portability is hard to achieve. Throughout the years, this trend resulted in an increasingly high burden on application developers to fine-tune their workloads for each architecture. This work explores how hardware-dependent aspects such as locality/process/thread affinity affect performance in modern CPU architectures. We focus our study on the Global Memory and Threading (GMT) distributed runtime system as a representative of Partitioned Global Address Space (PGAS) software stacks commonly adopted for productivity. In particular, to appreciate performance implications, we evaluate GMT’s thread affinity policies, and, introduce two new ones which exploit architectural awareness. Finally, we explore alternative NUMA configurations via different process bindings and perform a scalability study on three HPC clusters with varying CPU architectures and NUMA layouts. Our analysis indicates that more complex architectures are more affected by affinity and binding policies and highlights the importance of setting proper runtime configurations to achieve superior performance.

Di Dio Lavore, Ian↗

Compressed basis GMRES on high-performance graphics processing units

Krylov methods provide a fast and highly parallel numerical tool for the iterative solution of many large-scale sparse linear systems. To a large extent, the performance of practical realizations of these methods is constrained by the communication bandwidth in current computer architectures, motivating the investigation of sophisticated techniques to avoid, reduce, and/or hide the message-passing costs (in distributed platforms) and the memory accesses (in all architectures). This article leverages Ginkgo’s memory accessor in order to integrate a communication-reduction strategy into the (Krylov) GMRES solver that decouples the storage format (i.e., the data representation in memory) of the orthogonal basis from the arithmetic precision that is employed during the operations with that basis. Given that the execution time of the GMRES solver is largely determined by the memory accesses, the cost of the datatype transforms can be mostly hidden, resulting in the acceleration of the iterative step via a decrease in the volume of bits being retrieved from memory. Together with the special properties of the orthonormal basis (whose elements are all bounded by 1), this paves the road toward the aggressive customization of the storage format, which includes some floating-point as well as fixed-point formats with mild impact on the convergence of the iterative process. We develop a high-performance implementation of the “compressed basis GMRES” solver in the Ginkgo sparse linear algebra library using a large set of test problems from the SuiteSparse Matrix Collection. We demonstrate robustness and performance advantages on a modern NVIDIA V100 graphics processing unit (GPU) of up to 50% over the standard GMRES solver that stores all data in IEEE double-precision.

97 MATHEMATICS AND COMPUTING↗

Adaptive Mesh Refinement in the Fastlane

Adaptive Mesh Refinement (AMR) allows us to dramatically reduce memory usage by dynamically distributing resources to areas of physical significance.

97 MATHEMATICS AND COMPUTING↗

Optimizing the Energy Usage and Cognitive Value of Extreme-Scale Data-Analysis Approaches (Final Report)

This report serves as a record that we have successfully completed work on the proposal Optimizing the Energy Usage and Cognitive Value of Extreme Scale Analysis Approaches, or, more succinctly, ECX. This includes successful completion of all three phases of the project, findings as summarized below, strong publications across the fields represented in the proposal, and a set of artifacts (code, websites, and other documentation) available to the public. In total, the results of this project represent significant findings in each domain, and a toolkit that will allow the community to engage with this work, advancing it and contributing to it in the years ahead. They also form the basis for a successfully funded follow-on project. This executive summary presents findings, tools and publications resulting from the ECX project. The balance of the report then provides more detail on these results.

97 MATHEMATICS AND COMPUTING↗

Locating node of named data elements in coordination namespace

An approach is disclosed that locates a named data element by a local node. A name corresponding to the named data element is received, the named data element exists in a Coordination Namespace allocated in a memory area that is distributed amongst a set of nodes that include the local node and remote nodes. A predicted node identifier is received and then the named data element is requested from the predicted node based on the predicted node identifier.

97 MATHEMATICS AND COMPUTING↗

Fast entanglement distribution with atomic ensembles and fluorescent detection

Quantum repeaters based on atomic ensemble quantum memories are promising candidates for achieving scalable distribution of entanglement over long distances. Recently, important experimental progress has been made toward their implementation. However, the entanglement rates and scalability of current approaches are limited by relatively low retrieval and single-photon detector efficiencies. We propose a scheme which makes use of fluorescent detection of stored excitations to significantly increase the efficiency of connection and hence the rate. Practical performance and possible experimental realizations of the new protocol are discussed.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

NUMA-Aware Thread Scheduling for Big Data Transfers over Terabits Network Infrastructure

The evergrowing trend of big data has led scientists to share and transfer the simulation and analytical data across the geodistributed research and computing facilities. However, the existing data transfer frameworks used for data sharing lack the capability to adopt the attributes of the underlying parallel file systems (PFS). LADS (Layout-Aware Data Scheduling) is an end-to-end data transfer tool optimized for terabit network using a layout-aware data scheduling via PFS. However, it does not consider the NUMA (Nonuniform Memory Access) architecture. In this paper, we propose a NUMA-aware thread and resource scheduling for optimized data transfer in terabit network. First, we propose distributed RMA buffers to reduce memory controller contention in CPU sockets and then schedule the threads based on CPU socket and NUMA nodes inside CPU socket to reduce memory access latency. We design and implement the proposed resource and thread scheduling in the existing LADS framework. Experimental results showed from 21.7% to 44% improvement with memory-level optimizations in the LADS framework as compared to the baseline without any optimization.

97 MATHEMATICS AND COMPUTING↗

On the memory attribution problem: A solution and case study using MPI

As parallel applications running on large–scale computing systems become increasingly memory constrained, the ability to attribute memory usage to the various components of the application is becoming increasingly important. We present the design and implementation of memnesia, a novel memory usage profiler for parallel and distributed message–passing applications. Our approach captures both application– and message–passing library–specific memory usage statistics from unmodified binaries dynamically linked to a message–passing communication library. Using microbenchmarks and proxy applications, we evaluated our profiler across three Message Passing Interface (MPI) implementations and two hardware platforms. Furthermore, the results show that our approach and the corresponding implementation can accurately quantify memory resource usage as a function of time, scale, communication workload, and software or hardware system architecture, clearly distinguishing between application and MPI library memory usage at a per–process level. With this new capability, we show that job size, communication workload, and hardware/software architecture influence peak runtime memory usage. In practice, this tool provides a potentially valuable source of information for application developers seeking to measure and optimize memory usage.

97 MATHEMATICS AND COMPUTING↗

Cosmic string loop distribution on all length scales and at any redshift

We analytically derive the expected number density distribution of Nambu-Goto cosmic string loops at any redshift soon after the time of string formation to today. Our approach is based on the Polchinski-Rocha model of loop formation from long strings which we adjust to fit numerical simulations and complement by a phenomenological modelling of gravitational backreaction. Cosmological evolution drives the loop distribution towards scaling on all length scales in both the radiation and matter era. Memory of any reasonable initial loop distribution in the radiation era is shown to be erased well before Big Bang Nucleosynthesis. In the matter era, the loop distribution reaches full scaling, up to some residual loops from the radiation era which may be present for extremely low string tension. Finally, the number density of loops below the gravitational cutoff is shown to be scale independent, proportional to a negative power of the string tension and insensitive to the details of the backreaction modelling. As an application, we show that the energy density parameter of loops today cannot exceed 10{sup −5} for currently allowed string tension values, while the loop number density cannot be less than 10{sup −6} per Mpc{sup 3}. Our result should provide a more robust basis for studying the cosmological consequences of cosmic string loops.

79 ASTRONOMY AND ASTROPHYSICS↗

ac Stark gradient echo memory in cold atoms

The burgeoning fields of quantum computing and quantum key distribution have created a demand for a quantum memory. The gradient echo memory scheme is a quantum memory candidate for light storage that can boast efficiencies approaching unity, as well as the flexibility to work with either two- or three-level atoms. The key to this scheme is the frequency gradient that is placed across the memory. Currently, the three-level implementation uses a Zeeman gradient and warm atoms. In this article we model an alternate gradient-creation mechanism--the ac Stark effect--to provide an improvement in the flexibility of gradient-creation and field-switching times. We propose this scheme in concert with a move to cold atoms ({approx_equal}1 mK). These temperatures would increase the storage times possible, and the small ensemble volumes would enable large ac Stark shifts with reasonable laser power. We find that memory bandwidths on the order of MHz can be produced with experimentally achievable laser powers and trapping volumes, with high precision in gradient creation and switching times on the order of nanoseconds possible. By looking at the different decoherence mechanisms present in this system, we determine that coherence times on the order of tens of milliseconds are possible, as are delay-bandwidth products of approximately 50 and efficiencies over 90%.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Effect of local surface potential distribution on its relaxation in polycrystalline ferroelectric films.

We have studied the effect of local surface potential distribution on its relaxation in the polycrystalline ferroelectric thin films. A lower surface potential region, i.e., potential pit, is generated near a grain boundary. The deep potential pit has a faster relaxation than the area far away from the potential pit due to the acceleration of the screen charge draining near the grain boundary and the domains formed by applying higher voltage have a faster relaxation due to the larger gradient of screen charge distribution. In addition, the surface potential and its relaxation depend on the sign of applying voltage. The result shows that the surface potential distribution may influence significantly to the reliability of bit signal on the memory devices.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Multiple-stable anisotropic magnetoresistance memory in antiferromagnetic MnTe

Commercial magnetic memories rely on the bistability of ordered spins in ferromagnetic materials. Recently, experimental bistable memories have been realized using fully compensated antiferromagnetic metals. Here we demonstrate a multiple-stable memory device in epitaxial MnTe, an antiferromagnetic counterpart of common II–VI semiconductors. Favourable micromagnetic characteristics of MnTe allow us to demonstrate a smoothly varying zero-field antiferromagnetic anisotropic magnetoresistance (AMR) with a harmonic angular dependence on the writing magnetic field angle, analogous to ferromagnets. The continuously varying AMR provides means for the electrical read-out of multiple-stable antiferromagnetic memory states, which we set by heat-assisted magneto-recording and by changing the writing field direction. The multiple stability in our memory is ascribed to different distributions of domains with the Ne´el vector aligned along one of the three magnetic easy axes. The robustness against strong magnetic field perturbations combined with the multiple stability of the magnetic memory states are unique properties of antiferromagnets.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗