Search NASA⌕ Search

SEARCH · Search NASA

Results for “memory efficiency”

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 973 records · Page 54

Particle Communication and Domain Neighbor Coupling: Scalable Domain Decomposed Algorithms for Monte Carlo Particle Transport

In order to run Monte Carlo particle transport calculations on new supercomputers with hundreds of thousands or millions of processors, care must be taken to implement scalable algorithms. This means that the algorithms must continue to perform well as the processor count increases. In this paper, we examine the scalability of:(1) globally resolving the particle locations on the correct processor, (2) deciding that particle streaming communication has finished, and (3) efficiently coupling neighbor domains together with different replication levels. We have run domain decomposed Monte Carlo particle transport on up to 2 21 = 2,097,152 MPI processes on the IBM BG/Q Sequoia supercomputer and observed scalable results that agree with our theoretical predictions. These calculations were carefully constructed to have the same amount of work on every processor, i.e. the calculation is already load balanced. We also examine load imbalanced calculations where each domain’s replication level is proportional to its particle workload. In this case we show how to efficiently couple together adjacent domains to maintain within workgroup load balance and minimize memory usage.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Investigating High Performance RMA Interfaces for the MPI-3 Standard

The MPI-2 Standard, released in 1997, defined an interface for one-sided communication, also known as remote memory access (RMA). It was designed with the goal that it should permit efficient implementations on multiple platforms and networking technologies, and also in heterogeneous environments and non-cache-coherent systems. Nonetheless, even 12 years after its existence, the MPI-2 RMA interface remains scarcely used for a number of reasons. This paper discusses the limitations of the MPI-2 RMA specification, outlines the goals and requirements for a new RMA API that would better meet the needs of both users and implementers, and presents a strawman proposal for such an API. We also study the tradeoffs facing the design of this new API and discuss how it may be implemented efficiently on both cache-coherent and non-cache-coherent systems.

97 MATHEMATICS AND COMPUTING↗

Processing of Ni30Pt20Ti50 High-Temperature Shape-Memory Alloy Into Thin Rod Demonstrated

High-temperature shape-memory alloys (HTSMAs) based on nickel-titanium (NiTi) with significant ternary additions of palladium (Pd), platinum (Pt), gold (Au), or hafnium (Hf) have been identified as potential high-temperature actuator materials for use up to 500 C. These materials provide an enabling technology for the development of "smart structures" used to control the noise, emissions, or efficiency of gas turbine engines. The demand for these high-temperature versions of conventional shape-memory alloys also has been growing in the automotive, process control, and energy industries. However these materials, including the NiPtTi alloys being developed at the NASA Glenn Research Center, will never find widespread acceptance unless they can be readily processed into useable forms.

Noebe, Ronald D.↗

DIRAQ: scalable in situ data- and resource-aware indexing for optimized query performance

Scientific data analytics in high-performance computing environments has been evolving along with the advancement of computing capabilities. With the onset of exascale computing, the increasing gap between compute performance and I/O bandwidth has rendered the traditional post-simulation processing a tedious process. Despite the challenges due to increased data production, there exists an opportunity to benefit from "cheap" computing power to perform query-driven exploration and visualization during simulation time. To accelerate such analyses, applications traditionally augment, post-simulation, raw data with large indexes, which are then repeatedly utilized for data exploration. However, the generation of current state-of-the-art indexes involves a compute- and memory-intensive processing, thus rendering them inapplicable in an in situ context. In this paper we propose DIRAQ, a parallel in situ, in network data encoding and reorganization technique that enables the transformation of simulation output into a query-efficient form, with negligible runtime overhead to the simulation run. DIRAQ's effective core-local, precision-based encoding approach incorporates an embedded compressed index that is 3---6 $$\times $$ ? smaller than current state-of-the-art indexing schemes. Its data-aware index adjustmentation improves performance of group-level index layout creation by up to 35 % and reduces the size of the generated index by up to 27 %. Moreover, DIRAQ's in network index merging strategy enables the creation of aggregated indexes that speed up spatial-context query responses by up to $$10\times $$ 10 ? versus alternative techniques. DIRAQ's topology-, data-, and memory-aware aggregation strategy results in efficient I/O and yields overall end-to-end encoding and I/O time that is less than that required to write the raw data with MPI collective I/O.

Computer Science↗

Development of a GPU-based high-performance radiative transfer model for the Infrared Atmospheric Sounding Interferometer (IASI)

Satellite-observed radiance is a nonlinear functional of surface properties and atmospheric temperature and absorbing gas profiles as described by the radiative transfer equation (RTE). In the era of hyperspectral sounders with thousands of high-resolution channels, the computation of the radiative transfer model becomes more time-consuming. The radiative transfer model performance in operational numerical weather prediction systems still limits the number of channels we can use in hyperspectral sounders to only a few hundreds. To take the full advantage of such high-resolution infrared observations, a computationally efficient radiative transfer model is needed to facilitate satellite data assimilation. In recent years the programmable commodity graphics processing unit (GPU) has evolved into a highly parallel, multi-threaded, many-core processor with tremendous computational speed and very high memory bandwidth. The radiative transfer model is very suitable for the GPU implementation to take advantage of the hardware's efficiency and parallelism where radiances of many channels can be calculated in parallel in GPUs. In this paper, we develop a GPU-based high-performance radiative transfer model for the Infrared Atmospheric Sounding Interferometer (IASI) launched in 2006 onboard the first European meteorological polar-orbiting satellites, METOP-A. Each IASI spectrum has 8461 spectral channels. The IASI radiative transfer model consists of three modules. The first module for computing the regression predictors takes less than 0.004% of CPU time, while the second module for transmittance computation and the third module for radiance computation take approximately 92.5% and 7.5%, respectively. Our GPU-based IASI radiative transfer model is developed to run on a low-cost personal supercomputer with four GPUs with total 960 compute cores, delivering near 4 TFlops theoretical peak performance. By massively parallelizing the second and third modules, we reached 364x speedup for 1 GPU and 1455x speedup for all 4 GPUs, both with respect to the original CPU-based single-threaded Fortran code with the -O{sub 2} compiling optimization. The significant 1455x speedup using a computer with four GPUs means that the proposed GPU-based high-performance forward model is able to compute one day's amount of 1,296,000 IASI spectra within nearly 10 min, whereas the original single CPU-based version will impractically take more than 10 days. This model runs over 80% of the theoretical memory bandwidth with asynchronous data transfer. A novel CPU-GPU pipeline implementation of the IASI radiative transfer model is proposed. The GPU-based high-performance IASI radiative transfer model is suitable for the assimilation of the IASI radiance observations into the operational numerical weather forecast model.

97 MATHEMATICS AND COMPUTING↗

Information coding of exciting laser pulses in an optical echo-processor

We report the possibility of controlling the distribution of quantum bits within an inhomogeneously broadened line of a resonant transition in recording and transforming information in optical echo-processors. We consider the efficiency of realisation of the elementary logic XOR gate based on a two-pulse excitation of a resonant medium with phase memory. The encoded information is incorporated into the temporal shape of laser pulses in the form of amplitude modulation of an 'echelon' of present ('1') and absent ('0') pulse-codes for obtaining more efficient logic elements that reduce the noise in a quantum communication channel. (optical information processing)

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Efficient use of direct solvers for the calculation of compressible flows

While the direct solution of systems of linear equations resulting from fluid dynamic problems has generally not been practical in the past, it is presently demonstrated that the direct method is often more efficient than the most popular iterative schemes when constructed in such a way as to take advantage of presently available vector processing capabilities and large memory. The vertical line Gauss-Seidel algorithm was chosen as the iterative method to be compared with the direct method. It is fond that the direct method becomes efficient only when large residual reductions are desired.

Riggins, David W.↗

Locality Aware Concurrent Start for Stencil Applications

Stencil computations are at the heart of many physical simulations used in scientific codes. Thus, there exists a plethora of optimization efforts for this family of computations. Among these techniques, tiling techniques that allow concurrent start have proven to be very efficient in providing better performance for these critical kernels. Nevertheless, with many core designs being the norm, these optimization techniques might not be able to fully exploit locality (both spatial and temporal) on multiple levels of the memory hierarchy without compromising parallelism. It is no longer true that the machine can be seen as a homogeneous collection of nodes with caches, main memory and an interconnect network. New architectural designs exhibit complex grouping of nodes, cores, threads, caches and memory connected by an ever evolving network-on-chip design. These new designs may benefit greatly from carefully crafted schedules and groupings that encourage parallel actors (i.e. threads, cores or nodes) to be aware of the computational history of other actors in close proximity. In this paper, we provide an efficient tiling technique that allows hierarchical concurrent start for memory hierarchy aware tile groups. Each execution schedule and tile shape exploit the available parallelism, load balance and locality present in the given applications. We demonstrate our technique on the Intel Xeon Phi architecture with selected and representative stencil kernels. We show improvement ranging from 5.58% to 31.17% over existing state-of-the-art techniques.

Locality Aware execution, jagged tiling↗

NVMalloc: Exposing an Aggregate SSD Store as a Memory Partition in Extreme-Scale Machines

DRAM is a precious resource in extreme-scale machines and is increasingly becoming scarce, mainly due to the growing number of cores per node. On future multi-petaflop and exaflop machines, the memory pressure is likely to be so severe that we need to rethink our memory usage models. Fortunately, the advent of non-volatile memory (NVM) offers a unique opportunity in this space. Current NVM offerings possess several desirable properties, such as low cost and power efficiency, but suffer from high latency and lifetime issues. We need rich techniques to be able to use them alongside DRAM. In this paper, we propose a novel approach for exploiting NVM as a secondary memory partition so that applications can explicitly allocate and manipulate memory regions therein. More specifically, we propose an NVMalloc library with a suite of services that enables applications to access a distributed NVM storage system. We have devised ways within NVMalloc so that the storage system, built from compute node-local NVM devices, can be accessed in a byte-addressable fashion using the memory mapped I/O interface. Our approach has the potential to re-energize out-of-core computations on large-scale machines by having applications allocate certain variables through NVMalloc, thereby increasing the overall memory capacity available. Our evaluation on a 128-core cluster shows that NVMalloc enables applications to compute problem sizes larger than the physical memory in a cost-effective manner. It can bring more performance/efficiency gain with increased computation time between NVM memory accesses or increased data access locality. In addition, our results suggest that while NVMalloc enables transparent access to NVM-resident variables, the explicit control it provides is crucial to optimize application performance.

Computer Science↗

Evaluating Awkward Arrays, uproot, and coffea as a query platform for High Energy Physics Data

Query languages for High Energy Physics (HEP) are an ever present topic within the field. A query language that can efficiently represent the nested data structures that encode the statistical and physical meaning of HEP data will help analysts by ensuring their code is more clear and pertinent. As the result of a multi-year effort to develop an in-memory columnar representation of high energy physics data, the NumPy, Awkward Array, and uproot Python packages present a mature and efficient interface to HEP data. Atop that base, the coffea package adds functionality to launch queries at scale, manage and apply experiment-specific transformations to data, and present a rich object-oriented columnar data representation to the analyst. Recently, a set of Analysis Description Language (ADL) benchmarks has been established to compare HEP queries in multiple languages and frameworks. In this paper we present these benchmark queries implemented within the coffea framework and discuss their readability and performance characteristics. We find that the columnar queries perform as well or better than the implementations given in previous studies.

Gray, L.↗

Low-Storage, Explicit Runge-Kutta Schemes for the Compressible Navier-Stokes Equations

The derivation of storage explicit Runge-Kutta (ERK) schemes has been performed in the context of integrating the compressible Navier-Stokes equations via direct numerical simulation. Optimization of ERK methods is done across the broad range of properties, such as stability and accuracy efficiency, linear and nonlinear stability, error control reliability, step change stability, and dissipation/dispersion accuracy, subject to varying degrees of memory economization. Following van der Houwen and Wray, 16 ERK pairs are presented using from two to five registers of memory per equation, per grid point and having accuracies from third- to fifth-order. Methods have been assessed using the differential equation testing code DETEST, and with the 1D wave equation. Two of the methods have been applied to the DNS of a compressible jet as well as methane-air and hydrogen-air flames. Derived 3(2) and 4(3) pairs are competitive with existing full-storage methods. Although a substantial efficiency penalty accompanies use of two- and three-register, fifth-order methods, the best contemporary full-storage methods can be pearl), matched while still saving two to three registers of memory.

Kennedy, Chistopher A.↗

The Effects of Task Structure on Time-sharing Efficiency and Resource Allocation Optimality

A distinction was made between two aspects of time sharing performance: time sharing efficiency and attention allocation optimality. A secondary task technique was employed to evaluate the effects of the task structures of the component time shared tasks on both aspects of the time sharing performance. Five pairs of dual tasks differing in their structural configurations were investigated. The primary task was a visual/manual tracking task which requires spatial processing. The secondary task was either another tracking task or a verbal memory task with one of four different input/output configurations. Congruent to a common finding, time-sharing efficiency was observed to decrease with an increasing overlap of resources utilized by the time shared tasks. Research also tends to support the hypothesis that resource allocation is more optimal when the time shared tasks placed heavy demands on common processing resources than when they utilized separate resources.

Tsang, P. S.↗

A time-dependent neutron transport method of characteristics formulation with time derivative propagation

A new time-dependent Method of Characteristics (MOC) formulation for nuclear reactor kinetics was developed utilizing angular flux time-derivative propagation. This method avoids the requirement of storing the angular flux at previous points in time to represent a discretized time derivative; instead, an equation for the angular flux time derivative along 1D spatial characteristics is derived and solved concurrently with the 1D transport characteristic equation. This approach allows the angular flux time derivative to be recast principally in terms of the neutron source time derivatives, which are approximated to high-order accuracy using the backward differentiation formula (BDF). This approach, called Source Derivative Propagation (SDP), drastically reduces the memory requirements of time-dependent MOC relative to methods that require storing the angular flux. An SDP method was developed for 2D and 3D applications and implemented in the computer code DeCART in 2D. DeCART was used to model two reactor transient benchmarks: a modified TWIGL problem and a C5G7 transient. The SDP method accurately and efficiently replicated the solution of the conventional time-dependent MOC method using two orders of magnitude less memory.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Sparse Symmetric Format for Tucker Decomposition

Tensor-based methods are receiving renewed attention in recent years due to their prevalence in diverse real-world applications. There is considerable literature on tensor representations and algorithms for tensor decompositions, both for dense and sparse tensors. Many applications in hypergraph analytics, machine learning, psychometry, and signal processing result in tensors that are both sparse and symmetric, making them an important class for further study. Similar to the critical Tensor Times Matrix chain operation (TTM c ) in general sparse tensors, the $\underline{S}$ parse $\underline{S}$ ymmetric $\underline{T}$ ensor $\underline{T}$ imes $\underline{S}$ ame $\underline{M}$ atrix $\underline{c}$ hain (S 3 TTM c ) operation is compute and memory intensive due to high tensor order and the associated factorial explosion in the number of non-zeros. We present the novel Compressed Sparse Symmetric (CSS) format for sparse symmetric tensors, along with an efficient parallel algorithm for the S 3 TTM c operation. We theoretically establish that S 3 TTM c on CSS achieves a better memory versus run-time trade-off compared to state-of-the-art implementations, and visualize the variation of the performance gap over the parameter space. We demonstrate experimental findings that confirm these results and achieve up to 2.72× speedup on synthetic and real datasets. The scaling of the algorithm on different test architectures is also showcased to highlight the effect of machine characteristics on algorithm performance.

42 ENGINEERING↗

Querying Large Scientific Data Sets with Adaptable IO System ADIOS

When working with a large dataset, a relatively small fraction of data records are of interest in each analysis operation. For example, while examining a billion-particle dataset from an accelerator model, the scientists might focus on a few thousand fastest particles, or on the particle farthest from the beam center. In general, this type of selective data access is challenging because the selected data records could be anywhere in the dataset and require a significant amount of time to locate and retrieve. In this paper, we report our experience of addressing this data access challenge with the Adaptable IO System ADIOS. More specifically, we design a query interface for ADIOS to allow arbitrary combinations of range conditions on known variables, implement a number of different mechanisms for resolving these selection conditions, and devise strategies to reduce the time needed to retrieve the scattered data records. In many cases, the query mechanism can retrieve the selected data records orders of magnitude faster than the brute-force approach.Our work relies heavily on the in situ data processing feature of ADIOS to allow user functions to be executed in the data transport pipeline. This feature allows us to build indexes for efficient query processing, and to perform other intricate analyses while the data is in memory.

Gu, Junmin↗

A scale-wise analysis of intermittent momentum transport in dense canopy flows

We investigate the intermittent dynamics of momentum transport and its underlying time scales in the near-wall region of the neutrally stratified atmospheric boundary layer in the presence of a vegetation canopy. This is achieved through an empirical analysis of the persistence time scales (periods between successive zero-crossings) of momentum flux events, and their connection to the ejection–sweep cycle. Using high-frequency measurements from the GoAmazon campaign, spanning multiple heights within and above a dense canopy, the analysis suggests that, when the persistence time scales ( $t_p$ ) of momentum flux events from four different quadrants are separately normalized by $\varGamma _{w}$ (integral time scale of the vertical velocity), their distributions $P(t_p/\varGamma _{w})$ remain height-invariant. This result points to a persistent memory imposed by canopy-induced coherent structures, and to their role as an efficient momentum-transporting mechanism between the canopy airspace and the region immediately above. Moreover, $P(t_p/\varGamma _{w})$ exhibits a power-law scaling at times $t_{p}<\varGamma _{w}$ , with an exponential tail appearing for $t_{p} \geq \varGamma _{w}$ . By separating the flux events based on $t_p$ , we discover that around 80 % of the momentum is transported through the long-lived events ( $t_{p} \geq \varGamma _{w}$ ) at heights immediately above the canopy, while the short-lived ones ( $t_{p} < \varGamma _{w}$ ) only contribute marginally ( $\approx 20\,\%$ ). To explain the role of instantaneous flux amplitudes in momentum transport, we compare the measurements with newly developed surrogate data and establish that the range of time scales involved with amplitude variations in the fluxes tends to increase as one transitions from within to above the canopy.

Mechanics↗

Impact of Linearity and Write Noise of Analog Resistive Memory Devices in a Neural Algorithm Accelerator

Resistive memory (ReRAM) shows promise for use as an analog synapse element in energy-efficient neural network algorithm accelerators. A particularly important application is the training of neural networks, as this is the most computationally-intensive procedure in using a neural algorithm. However, training a network with analog ReRAM synapses can significantly reduce the accuracy at the algorithm level. In order to assess this degradation, analog properties of ReRAM devices were measured and hand-written digit recognition accuracy was modeled for the training using backpropagation. Bipolar filamentary devices utilizing three material systems were measured and compared: one oxygen vacancy system, Ta-TaO x , and two conducting metallization systems, Cu-SiO 2 , and Ag/chalcogenide. Analog properties and conductance ranges of the devices are optimized by measuring the response to varying voltage pulse characteristics. Key analog device properties which degrade the accuracy are update linearity and write noise. Write noise may improve as a function of device manufacturing maturity, but write nonlinearity appears relatively consistent among the different device material systems and is found to be the most significant factor affecting accuracy. As a result, this suggests that new materials and/or fundamentally different resistive switching mechanisms may be required to improve device linearity and achieve higher algorithm training accuracy.

97 MATHEMATICS AND COMPUTING↗

Integrated Model, Batch and Domain Parallelism in Training Neural Networks

We propose a new integrated method of exploiting model, batch and domain parallelism for the training of deep neural networks (DNNs) on large distributed-memory computers using minibatch stochastic gradient descent (SGD). Our goal is to find an efficient parallelization strategy for a fixed batch size using $P$ processes. Our method is inspired by the communication-avoiding algorithms in numerical linear algebra. We see $P$ processes as logically divided into a $P_r \times P_c$ grid where the $P_r$ dimension is implicitly responsible for model/domain parallelism and the $P_c$ dimension is implicitly responsible for batch parallelism. In practice, the integrated matrix-based parallel algorithm encapsulates these types of parallelism automatically. We analyze the communication complexity and analytically demonstrate that the lowest communication costs are often achieved neither with pure model nor with pure data parallelism. We also show how the domain parallel approach can help in extending the theoretical scaling limit of the typical batch parallel method.

97 MATHEMATICS AND COMPUTING↗