Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallelization 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 1,765 records · Page 98

A direct-execution parallel architecture for the Advanced Continuous Simulation Language (ACSL)

A direct-execution parallel architecture for the Advanced Continuous Simulation Language (ACSL) is presented which overcomes the traditional disadvantages of simulations executed on a digital computer. The incorporation of parallel processing allows the mapping of simulations into a digital computer to be done in the same inherently parallel manner as they are currently mapped onto an analog computer. The direct-execution format maximizes the efficiency of the executed code since the need for a high level language compiler is eliminated. Resolution is greatly increased over that which is available with an analog computer without the sacrifice in execution speed normally expected with digitial computer simulations. Although this report covers all aspects of the new architecture, key emphasis is placed on the processing element configuration and the microprogramming of the ACLS constructs. The execution times for all ACLS constructs are computed using a model of a processing element based on the AMD 29000 CPU and the AMD 29027 FPU. The increase in execution speed provided by parallel processing is exemplified by comparing the derived execution times of two ACSL programs with the execution times for the same programs executed on a similar sequential architecture.

Carroll, Chester C.↗

Multi-task Parallelism for Robust Pre-training of Graph Foundation Models on Multi-source, Multi-fidelity Atomistic Modeling Data

Graph foundation models using graph neural networks promise sustainable, efficient atomistic modeling. To tackle challenges of processing multi-source, multi-fidelity data during pre-training, recent studies employ multi-task learning, in which shared message passing layers initially process input atomistic structures regardless of source, then route them to multiple decoding heads that predict data-specific outputs. This approach stabilizes pre-training and enhances a model’s transferability to unexplored chemical regions. Preliminary results on approximately four million structures are encouraging, yet questions remain about generalizability to larger, more diverse datasets and scalability on supercomputers. We propose a multi-task parallelism method that distributes each head across computing resources with GPU acceleration. Implemented in the open-source HydraGNN architecture, our method was trained on over 24 million structures from five datasets and tested on the Perlmutter, Aurora, and Frontier supercomputers, demonstrating efficient scaling on all three highly heterogeneous super-computing architectures.

Lupo Pasini, Massimiliano [ORNL] (ORCID:0000000249↗

Active Flash: Out-of-core Data Analytics on Flash Storage

Next generation science will increasingly come to rely on the ability to perform efficient, on-the-fly analytics of data generated by high-performance computing (HPC) simulations, modeling complex physical phenomena. Scientific computing workflows are stymied by the traditional chaining of simulation and data analysis, creating multiple rounds of redundant reads and writes to the storage system, which grows in cost with the ever-increasing gap between compute and storage speeds in HPC clusters. Recent HPC acquisitions have introduced compute node-local flash storage as a means to alleviate this I/O bottleneck. We propose a novel approach, Active Flash, to expedite data analysis pipelines by migrating to the location of the data, the flash device itself. We argue that Active Flash has the potential to enable true out-of-core data analytics by freeing up both the compute core and the associated main memory. By performing analysis locally, dependence on limited bandwidth to a central storage system is reduced, while allowing this analysis to proceed in parallel with the main application. In addition, offloading work from the host to the more power-efficient controller reduces peak system power usage, which is already in the megawatt range and poses a major barrier to HPC system scalability. We propose an architecture for Active Flash, explore energy and performance trade-offs in moving computation from host to storage, demonstrate the ability of appropriate embedded controllers to perform data analysis and reduction tasks at speeds sufficient for this application, and present a simulation study of Active Flash scheduling policies. These results show the viability of the Active Flash model, and its capability to potentially have a transformative impact on scientific data analysis.

99 GENERAL AND MISCELLANEOUS↗

Parallel processing for scientific computations

The main contribution of the effort in the last two years is the introduction of the MOPPS system. After doing extensive literature search, we introduced the system which is described next. MOPPS employs a new solution to the problem of managing programs which solve scientific and engineering applications on a distributed processing environment. Autonomous computers cooperate efficiently in solving large scientific problems with this solution. MOPPS has the advantage of not assuming the presence of any particular network topology or configuration, computer architecture, or operating system. It imposes little overhead on network and processor resources while efficiently managing programs concurrently. The core of MOPPS is an intelligent program manager that builds a knowledge base of the execution performance of the parallel programs it is managing under various conditions. The manager applies this knowledge to improve the performance of future runs. The program manager learns from experience.

Alkhatib, Hasan S.↗

Genetic algorithm based task reordering to improve the performance of batch scheduled massively parallel scientific applications

The growth in size of networked high performance computers along with novel accelerator-based node architectures has further emphasized the importance of communication efficiency in high performance computing. The world's largest high performance computers are usually operated as shared user facilities due to the costs of acquisition and operation. Applications are scheduled for execution in a shared environment and are placed on nodes that are not necessarily contiguous on the interconnect. Furthermore, the placement of tasks on the nodes allocated by the scheduler is sub-optimal, leading to performance loss and variability. Here, we investigate the impact of task placement on the performance of two massively parallel application codes on the Titan supercomputer, a turbulent combustion flow solver (S3D) and a molecular dynamics code (LAMMPS). Benchmark studies show a significant deviation from ideal weak scaling and variability in performance. The inter-task communication distance was determined to be one of the significant contributors to the performance degradation and variability. A genetic algorithm-based parallel optimization technique was used to optimize the task ordering. This technique provides an improved placement of the tasks on the nodes, taking into account the application's communication topology and the system interconnect topology. As a result, application benchmarks after task reordering through genetic algorithm show a significant improvement in performance and reduction in variability, therefore enabling the applications to achieve better time to solution and scalability on Titan during production.

97 MATHEMATICS AND COMPUTING↗

A GPU‐Accelerated Generative Adversarial Model for Causal Inference

We develop a GPU-accelerated machine learning generative adversarial model designed to facilitate causal inferences from observational data. Our model's theoretical framework is conceptualized in a manner that is amenable to being operable and scalable for high-performance computing platforms. We leverage GPU acceleration to develop a parallel evolutionary algorithm to achieve large-scale parallel computation of the model within a now widely accessible computing platform. This capability both enhances computational speedup and efficiency and also extends the use of the model to a broader range of substantive research domains while maintaining the underlying theoretical properties of the model.

GPU↗

Implementing High Performance Remote Method Invocation in CCA

We report our effort in engineering a high performance remote method invocation (RMI) mechanism for the Common Component Architecture (CCA). This mechanism provides a highly efficient and easy-to-use mechanism for distributed computing in CCA, enabling CCA applications to effectively leverage parallel systems to accelerate computations. This work is built on the previous work of Babel RMI. Babel is a high performance language interoperability tool that is used in CCA for scientific application writers to share, reuse, and compose applications from software components written in different programming languages. Babel provides a transparent and flexible RMI framework for distributed computing. However, the existing Babel RMI implementation is built on top of TCP and does not provide the level of performance required to distribute fine-grained tasks. We observed that the main reason the TCP based RMI does not perform well is because it does not utilize the high performance interconnect hardware on a cluster efficiently. We have implemented a high performance RMI protocol, HPCRMI. HPCRMI achieves low latency by building on top of a low-level portable communication library, Aggregated Remote Message Copy Interface (ARMCI), and minimizing communication for each RMI call. Our design allows a RMI operation to be completed by only two RDMA operations. We also aggressively optimize our system to reduce copying. In this paper, we discuss the design and our experimental evaluation of this protocol. Our experimental results show that our protocol can improve RMI performance by an order of magnitude.

99 GENERAL AND MISCELLANEOUS↗

An algorithm for continuum modeling of rocks with multiple embedded nonlinearly-compliant joints [Continuum modeling of elasto-plastic media with multiple embedded nonlinearly-compliant joints]

Here, we present a numerical method for modeling the mechanical effects of nonlinearly-compliant joints in elasto-plastic media. The method uses a series of strain-rate and stress update algorithms to determine joint closure, slip, and solid stress within computational cells containing multiple “embedded” joints. This work facilitates efficient modeling of nonlinear wave propagation in large spatial domains containing a large number of joints that affect bulk mechanical properties. We implement the method within the massively parallel Lagrangian code GEODYN-L and provide verification and examples. We highlight the ability of our algorithms to capture joint interactions and multiple weakness planes within individual computational cells, as well as its computational efficiency. We also discuss the motivation for developing the proposed technique: to simulate large-scale wave propagation during the Source Physics Experiments (SPE), a series of underground explosions conducted at the Nevada National Security Site (NNSS).

42 ENGINEERING↗

Simulation of thermal hydraulic performance of multiple parallel micropin arrays for concentrating solar thermal applications with supercritical carbon dioxide

Concentrated solar power (CSP) plants have the potential to provide 24 h, renewable electricity. Current CSP systems have high capital and operational costs which makes the levelized cost of electricity uncompetitive with conventional techniques. Recent experimental research has shown the potential of small unit cells (up to 2 × 2 cm) containing micropin arrays (D H < 1 mm) to operate efficiently at high incident flux (>140 W cm -2 ) using supercritical carbon dioxide as the working fluid. Applying this technology to CSP systems would result in a smaller central receiver, which would reduce thermal losses, increase receiver efficiency and reduce the capital cost of the receiver component. This study investigates and addresses the practical thermal and hydraulic issues in numbering up these small unit cells into numerous parallel cells within an integrated module design. A thermal hydraulic network model is developed to quantify the distribution and the overall receiver efficiency of an integrated module. This model is used to specify maximum allowable unit cell size and header dimensions to maintain acceptable thermal performance and pressure loss. Once a module design was finalized, parametric studies were performed to study the effects of varying incident flux on flow distribution and thermal performance of the module. Finally, the results show that an integrated module design can be achieved with less than 5% flow maldistribution and a pressure drop acceptable to the remainder of the system.

14 SOLAR ENERGY↗

Computing with Spikes: The Advantage of Fine-Grained Timing

Neural-inspired spike-based computing machines often claim to achieve considerable advantages in terms of energy and time efficiency by using spikes for computation and communication. However, fundamental questions about spike-based computation remain unanswered. For instance, how much advantage do spike-based approaches have over conventional methods, and under what circumstances does spike-based computing provide a comparative advantage? Simply implementing existing algorithms using spikes as the medium of computation and communication is not guaranteed to yield an advantage. Here, we demonstrate that spike-based communication and computation within algorithms can increase throughput, and they can decrease energy cost in some cases. We present several spiking algorithms, including sorting a set of numbers in ascending/descending order, as well as finding the maximum or minimum or median of a set of numbers. We also provide an example application: a spiking median-filtering approach for image processing providing a low-energy, parallel implementation. Furthermore, the algorithms and analyses presented here demonstrate that spiking algorithms can provide performance advantages and offer efficient computation of fundamental operations useful in more complex algorithms.

97 MATHEMATICS AND COMPUTING↗

Unbalanced Parallel I/O: An Often-Neglected Side Effect of Lossy Scientific Data Compression

Lossy compression techniques have demonstrated promising results in significantly reducing the scientific data size while guaranteeing the compression error bounds. However, one important yet often neglected side effect of lossy scientific data compression is its impact on the performance of parallel I/O. Our key observation is that the compressed data size is often highly skewed across processes in lossy scientific compression. To understand this behavior, we conduct extensive experiments where we apply three lossy compressors MGARD, ZFP, and SZ, which are specifically designed and optimized for scientific data, to three real-world scientific applications Gray-Scott simulation, WarpX, and XGC. Our analysis result demonstrates that the size of the compressed data is always skewed even if the original data is evenly decomposed among processes. Such skewness widely exists in different scientific applications using different compressors as long as the information density of the data varies across processes. We then systematically study how this side effect of lossy scientific data compression impacts the performance of parallel I/O. We observe that the skewness in the sizes of the compressed data often leads to I/O imbalance, which can significantly reduce the efficiency of I/O bandwidth utilization if not properly handled. In addition, writing data concurrently to a single shared file through MPI-IO library is more sensitive to the unbalanced I/O loads. Therefore, we believe our research community should pay more attention to the unbalanced parallel I/O caused by lossy scientific data compression.

Wang, Xinying↗

Efficient and Effective Sparse Tensor Reordering

This paper proposed two reordering schemes for sparse tensors: BFS-MCS and LEXI-ORDER. BFS-MCS is a Breadth First Search (BFS)-like heuristic approach based on the maximum cardinality search family; LEXI-ORDER is an extension of doubly lexical ordering of matrices to tensors. CANDECOMP/PARAFAC decomposition (CPD) is taken as an example to show their effect on existing three sparse tensor formats for CPUs: coordinate (COO), compressed sparse fiber (CSF), and hierarchical coordinate (HICOO). LEXI-ORDER obtains up to 4.14× speedup on sequential HICOO-MTTKRP and 11.88× speedup on its parallel case. COO- and CSF-MTTKRP also achieves performance improvement in different degree. Our two reordering methods are more efficient and effective than other state-of-the-art reordering methods used in SPLATT.

HPC↗

SWAP-Assembler 2: Optimization of De Novo Genome Assembler at Large Scale

In this paper, we analyze and optimize the most time-consuming steps of the SWAP-Assembler, a parallel genome assembler, so that it can scale to a large number of cores for huge genomes with the size of sequencing data ranging from terabyes to petabytes. According to the performance analysis results, the most time-consuming steps are input parallelization, k-mer graph construction, and graph simplification (edge merging). For the input parallelization, the input data is divided into virtual fragments with nearly equal size, and the start position and end position of each fragment are automatically separated at the beginning of the reads. In k-mer graph construction, in order to improve the communication efficiency, the message size is kept constant between any two processes by proportionally increasing the number of nucleotides to the number of processes in the input parallelization step for each round. The memory usage is also decreased because only a small part of the input data is processed in each round. With graph simplification, the communication protocol reduces the number of communication loops from four to two loops and decreases the idle communication time. The optimized assembler is denoted as SWAP-Assembler 2 (SWAP2). In our experiments using a 1000 Genomes project dataset of 4 terabytes (the largest dataset ever used for assembling) on the supercomputer Mira, the results show that SWAP2 scales to 131,072 cores with an efficiency of 40%. We also compared our work with both the HipMER assembler and the SWAP-Assembler. On the Yanhuang dataset of 300 gigabytes, SWAP2 shows a 3X speedup and 4X better scalability compared with the HipMer assembler and is 45 times faster than the SWAP-Assembler. The SWAP2 software is available at https://sourceforge.net/projects/swapassembler.

Meng, Jintao↗

Diode-pumped laser research

The Laboratory for Oceans is currently working on the development of compact laser diode array (LD) pumped Nd:YAG lasers for use in space-based altimetry and ranging. Laser diode-array pumping technology promises to increase the electrical to optical efficiency of solid state lasers by an order of magnitude with a lifetime increase of nearly three orders of magnitude relative to today's conventional flashlamp-pumped laser systems. The small size, efficiency, and ruggedness make LD-pumped solid state lasers ideal for space based applications. In an in-house RTOP effort, a novel multiple-pass LD-pumped Nd:YAG laser amplifier was designed and tested to increase the 100 microjoule output pulse energy of the Lightwave laser oscillator. Preliminary results have yielded a round trip amplifier gain of about 15 percent using 7 microjoule LD-pump energy. As a parallel activity, funding was recently obtained to investigate the possible use of custom made fiber optic arrays to obtain an efficient optical coupling mechanism between the emitting laser diode-arrays and the target solid state laser material. Fiber optic coupling arrays would allow for the easy manipulation of the spatial emitting pattern of the diode pump sources to match either an end or side pumping laser configuration.

Ramos-Izquierdo, L.↗

An Efficient and Accurate Algorithm for Computing Grid-Averaged Solar Fluxes for Horizontally Inhomogeneous Clouds

A computationally efficient method is presented to account for the horizontal cloud inhomogeneity by using a radiatively equivalent plane parallel homogeneous (PPH) cloud. The algorithm can accurately match the calculations of the reference (rPPH) independent column approximation (ICA) results, but use only the same computational time required for a single plane parallel computation. The effective optical depth of this synthetic sPPH cloud is derived by exactly matching the direct transmission to that of the inhomogeneous ICA cloud. The ffective9 scattering asymmetry factor is found from a pre-calculated albedo inverse look-up-table that is allowed to vary over the range from -1.0 to 1.0. In the special cases of conservative scattering and total absorption, the synthetic method is exactly equivalent to the ICA, with only a small bias (about 0.2% in flux) relative to ICA due to imperfect interpolation in using the look-up tables. In principle, the ICA albedo can be approximated accurately regardless of cloud inhomogeneity. For a more complete comparison, the broadband shortwave albedo and transmission calculated from the synthetic sPPH cloud and averaged over all incident directions, have the RMS biases of 0.26% and 0.76%, respectively, for inhomogeneous clouds over a wide variation of particle size. The advantages of the synthetic PPH method are that (1) it is not required that all the cloud subcolumns have uniform microphysical characteristic, (2) it is applicable to any 1D radiative transfer scheme, and (3) it can handle arbitrary cloud optical depth distributions and an arbitrary number of cloud subcolumns with uniform computational efficiency.

cloud inhomogeneity↗

Parallel Calculation of Sensitivity Derivatives for Aircraft Design using Automatic Differentiation

Sensitivity derivative (SD) calculation via automatic differentiation (AD) typical of that required for the aerodynamic design of a transport-type aircraft is considered. Two ways of computing SD via code generated by the ADIFOR automatic differentiation tool are compared for efficiency and applicability to problems involving large numbers of design variables. A vector implementation on a Cray Y-MP computer is compared with a coarse-grained parallel implementation on an IBM SP1 computer, employing a Fortran M wrapper. The SD are computed for a swept transport wing in turbulent, transonic flow; the number of geometric design variables varies from 1 to 60 with coupling between a wing grid generation program and a state-of-the-art, 3-D computational fluid dynamics program, both augmented for derivative computation via AD. For a small number of design variables, the Cray Y-MP implementation is much faster. As the number of design variables grows, however, the IBM SP1 becomes an attractive alternative in terms of compute speed, job turnaround time, and total memory available for solutions with large numbers of design variables. The coarse-grained parallel implementation also can be moved easily to a network of workstations.

Bischof, c. H.↗

Synthetic-domain computing and neural networks using lithium niobate integrated nonlinear phononics

Analogue computing uses the physical behaviours of devices to provide energy-efficient arithmetic operations. However, scaling up analogue computing platforms by simply increasing the number of devices leads to challenges such as device-to-device variation. Here, in this study, we report scalable analogue computing and neural networks in the synthetic frequency domain using an integrated nonlinear phononic platform on lithium niobate. This synthetic-domain computing is robust to device variations, as vectors and matrices are concurrently encoded at different frequencies within a single device, achieving a high throughput per area. Leveraging inherent nonlinearities, our device-aware neural network can perform a four-class classification task with an accuracy of 98.2%. The nonlinear phononic computing hardware also maintains consistent performance over a wide operational temperature range (characterized up to 192 °C). Our synthetic-domain computing combines single-device parallelism, inherent nonlinearity and environmental stability, and could be of use in edge computing applications in which power efficiency and environmental resilience are crucial.

Ji, Jun [Virginia Polytechnic Inst. and State Univ↗

Optimal processor assignment for pipeline computations

The availability of large scale multitasked parallel architectures introduces the following processor assignment problem for pipelined computations. Given a set of tasks and their precedence constraints, along with their experimentally determined individual responses times for different processor sizes, find an assignment of processor to tasks. Two objectives are of interest: minimal response given a throughput requirement, and maximal throughput given a response time requirement. These assignment problems differ considerably from the classical mapping problem in which several tasks share a processor; instead, it is assumed that a large number of processors are to be assigned to a relatively small number of tasks. Efficient assignment algorithms were developed for different classes of task structures. For a p processor system and a series parallel precedence graph with n constituent tasks, an O(np2) algorithm is provided that finds the optimal assignment for the response time optimization problem; it was found that the assignment optimizing the constrained throughput in O(np2log p) time. Special cases of linear, independent, and tree graphs are also considered.

Nicol, David M.↗