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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

Pin-by-Pin Core Calculation with an NEM-Based Two-Level Hybrid CMFD Algorithm

In the near future, systematic pin-by-pin core calculations are expected to provide a high level of accuracy. These are being developed in pursuit of high-fidelity tools for analyzing the reactor configurations. One such tool that has recently gained significant attention is the pin-by-pin core calculation via local-global iteration using nonlinear acceleration scheme, such as the One-Node CMFD (Coarse-Mesh Finite Difference) nonlinear algorithm. The One-Node CMFD method uses two correction factors to preserve the interface current, in contrast to the one correction term employed by conventional nonlinear iterative methods such as that proposed by Kord. S. Smith. These correction factors are updated with surface flux and net current which are obtained from a high order local problem. With two correction factors, One-Node CMFD can reduce computing time of local FMFD (Fine-Mesh Finite Difference) as it nonlinearly couples the CMFD and FMFD methods in global-local iterations. Nonetheless, very small mesh size is still required and it consequently requires long computing time for high accuracy. As such, a more flexible approach is needed to address this challenge. This paper presents one such possible approach named the hCMFD (Hybrid CMFD) method. In this method, One-Node CMFD is used to solve the global problem, while Two-Node NEM (Nodal Expansion Method) CMFD replaces FMFD as the solver for the local problem. In this paper, two kinds of CMFD methods has been simultaneously implemented for a global-local nonlinear iteration in order to do pin-by-pin heterogeneous reactor calculations. The global problem is solved by a one-node CMFD which uses two correction factors, while the local fixed-source problem is solved by the conventional two-node NEM CMFD which use one correction factor on a surface. From solution of the global eigenvalue problem, incoming partial current is obtained and it is used as boundary condition of the local fixed-source problems. It has been shown that over 99% of the whole calculations can be done in parallel with the new CMFD method for 2-D problems and the algorithm can be easily parallelized by using the OpenMP parallel scheme. The hCMFD method will enable an efficient pin-by-pin reactor analysis since a very efficient parallel calculation is possible with the hCMFD method. In future works, the hCMFD method will be extended to three-dimensional reactor problem. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2011

Global Arrays Parallel Programming Toolkit

The two predominant classes of programming models for parallel computing are distributed memory and shared memory. Both shared memory and distributed memory models have advantages and shortcomings. Shared memory model is much easier to use but it ignores data locality/placement. Given the hierarchical nature of the memory subsystems in modern computers this characteristic can have a negative impact on performance and scalability. Careful code restructuring to increase data reuse and replacing fine grain load/stores with block access to shared data can address the problem and yield performance for shared memory that is competitive with message-passing. However, this performance comes at the cost of compromising the ease of use that the shared memory model advertises. Distributed memory models, such as message-passing or one-sided communication, offer performance and scalability but they are difficult to program. The Global Arrays toolkit attempts to offer the best features of both models. It implements a shared-memory programming model in which data locality is managed by the programmer. This management is achieved by calls to functions that transfer data between a global address space (a distributed array) and local storage. In this respect, the GA model has similarities to the distributed shared-memory models that provide an explicit acquire/release protocol. However, the GA model acknowledges that remote data is slower to access than local data and allows data locality to be specified by the programmer and hence managed. GA is related to the global address space languages such as UPC, Titanium, and, to a lesser extent, Co-Array Fortran. In addition, by providing a set of data-parallel operations, GA is also related to data-parallel languages such as HPF, ZPL, and Data Parallel C. However, the Global Array programming model is implemented as a library that works with most languages used for technical computing and does not rely on compiler technology for achieving parallel efficiency. It also supports a combination of task- and data-parallelism and is available as an extension of the message passing (MPI) model. The GA model exposes to the programmer the hierarchical memory of modern high-performance computer systems, and by recognizing the communication overhead for remote data transfer, it promotes data reuse and locality of reference. Virtually all the scalable architectures possess non-uniform memory access characteristics that reflect their multi-level memory hierarchies. These hierarchies typically comprise processor registers, multiple levels of cache, local memory, and remote memory. Over time, both the number of levels and the cost (in processor cycles) of accessing deeper levels has been increasing. It is important for any scalable programming model to address memory hierarchy since it is critical to the efficient execution of scalable applications.

ARMCI↗
DOE OSTI2019

CosmoFlow: Using Deep Learning to Learn the Universe at Scale

Deep learning is a promising tool to determine the physical model that describes our universe. To handle the considerable computational cost of this problem, we present CosmoFlow: a highly scalable deep learning application built on top of the TensorFlow framework. CosmoFlow uses efficient implementations of 3D convolution and pooling primitives, together with improvements in threading for many element-wise operations, to improve training performance on Intel ® Xeon Phi™ processors. We also utilize the Cray PE Machine Learning Plugin for efficient scaling to multiple nodes. We demonstrate fully synchronous data-parallel training on 8192 nodes of Cori with 77% parallel efficiency, achieving 3.5 Pflop/s sustained performance. To our knowledge, this is the first large-scale science application of the TensorFlow framework at supercomputer scale with fully-synchronous training. Here, these enhancements enable us to process large 3D dark matter distribution and predict the cosmological parameters Ω M , σ 8 and n s with unprecedented accuracy.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2019

A Lifeline-Based Approach for Work Requesting and Parallel Particle Advection

Particle advection, a fundamental building block for many flow visualization algorithms, is very difficult to parallelize efficiently. That said, work requesting is a promising technique to improve parallel performance for particle advection. With this work, we introduce a new work requesting-based method which uses the Lifeline scheduling method. To evaluate the impact of this new algorithm, we ran 92 experiments, running at concurrencies as high as 8192 cores, data sets as large as 17 billion cells, and as many as 16 million particles, comparing against other work requesting scheduling methods. Overall, our results show that Lifeline has significantly less idle time than other approaches, since it reduces the number of failed attempts to request work.

Pugmire, Dave↗
DOE OSTI2015

Scalable Parallel Methods for Analyzing Metagenomics Data at Extreme Scale

The field of bioinformatics and computational biology is currently experiencing a data revolution. The exciting prospect of making fundamental biological discoveries is fueling the rapid development and deployment of numerous cost-effective, high-throughput next-generation sequencing technologies. The result is that the DNA and protein sequence repositories are being bombarded with new sequence information. Databases are continuing to report a Moore’s law-like growth trajectory in their database sizes, roughly doubling every 18 months. In what seems to be a paradigm-shift, individual projects are now capable of generating billions of raw sequence data that need to be analyzed in the presence of already annotated sequence information. While it is clear that data-driven methods, such as sequencing homology detection, are becoming the mainstay in the field of computational life sciences, the algorithmic advancements essential for implementing complex data analytics at scale have mostly lagged behind. Sequence homology detection is central to a number of bioinformatics applications including genome sequencing and protein family characterization. Given millions of sequences, the goal is to identify all pairs of sequences that are highly similar (or “homologous”) on the basis of alignment criteria. While there are optimal alignment algorithms to compute pairwise homology, their deployment for large-scale is currently not feasible; instead, heuristic methods are used at the expense of quality. In this dissertation, we present the design and evaluation of a parallel implementation for conducting optimal homology detection on distributed memory supercomputers. Our approach uses a combination of techniques from asynchronous load balancing (viz. work stealing, dynamic task counters), data replication, and exact-matching filters to achieve homology detection at scale. Results for a collection of 2.56M sequences show parallel efficiencies of ~75-100% on up to 8K cores, representing a time-to-solution of 33 seconds. We extend this work with a detailed analysis of single-node sequence alignment performance using the latest CPU vector instruction set extensions. Preliminary results reveal that current sequence alignment algorithms are unable to fully utilize widening vector registers.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2019

Provenance–aware workflow for data quality management and improvement for large continuous scientific data streams

Data quality assessment, management and improvement is integral part of any big data intensive scientific research to ensure accurate, reliable and reproducible science discoveries. The task of maintaining the quality of data, however, is non-trivial and pose a challenge for a program like Department of Energy’s Atmospheric Radiation Measurement (ARM) that collects data from hundreds of instruments across the world,and distributes thousands of streaming data products that are continuously growing in near-real-time in an archive 1.7 Petabyte in size and growing. We present a computational data processing workflow to collect the data quality issue via an easy and intuitive web-based portal that allows reporting of any quality issues for any site, facility or instruments at a granularity down to individual variable in the data files. Portal allows instrument specialists and scientists to provide corrective actions in form of symbolic equation. A parallel processing framework applies the data improvement to large volume of data in efficient parallel environment, while optimizing data transfer and file I/O operations. Corrected files are systematically versioned and archived. A provenance tracking module tracks and records any change made to the data during its entire life cycle which are communicated transparently to the scientific users. Developed in Python using open source technologies, software architecture enables efficient and fast management and improvement of data in an operational data center environment.

Kumar, Jitendra↗
DOE OSTI2013

GROMACS 4.5: A high-throughput and highly parallel open source molecular simulation toolkit

In this study, molecular simulation has historically been a low-throughput technique, but faster computers and increasing amounts of genomic and structural data are changing this by enabling large-scale automated simulation of, for instance, many conformers or mutants of biomolecules with or without a range of ligands. At the same time, advances in performance and scaling now make it possible to model complex biomolecular interaction and function in a manner directly testable by experiment. These applications share a need for fast and efficient software that can be deployed on massive scale in clusters, web servers, distributed computing or cloud resources. As a result, we present a range of new simulation algorithms and features developed during the past 4 years, leading up to the GROMACS 4.5 software package. The software now automatically handles wide classes of biomolecules, such as proteins, nucleic acids and lipids, and comes with all commonly used force fields for these molecules built-in. GROMACS supports several implicit solvent models, as well as new free-energy algorithms, and the software now uses multithreading for efficient parallelization even on low-end systems, including windows-based workstations. Together with hand-tuned assembly kernels and state-of-the-art parallelization, this provides extremely high performance and cost efficiency for high-throughput as well as massively parallel simulations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2011

Parallel-vector algorithms for particle simulations on shared-memory multiprocessors

Over the last few decades, the computational demands of massive particle-based simulations for both scientific and industrial purposes have been continuously increasing. Hence, considerable efforts are being made to develop parallel computing techniques on various platforms. In such simulations, particles freely move within a given space, and so on a distributed-memory system, load balancing, i.e., assigning an equal number of particles to each processor, is not guaranteed. However, shared-memory systems achieve better load balancing for particle models, but suffer from the intrinsic drawback of memory access competition, particularly during (1) paring of contact candidates from among neighboring particles and (2) force summation for each particle. Here, novel algorithms are proposed to overcome these two problems. For the first problem, the key is a pre-conditioning process during which particle labels are sorted by a cell label in the domain to which the particles belong. Then, a list of contact candidates is constructed by pairing the sorted particle labels. For the latter problem, a table comprising the list indexes of the contact candidate pairs is created and used to sum the contact forces acting on each particle for all contacts according to Newton's third law. With just these methods, memory access competition is avoided without additional redundant procedures. The parallel efficiency and compatibility of these two algorithms were evaluated in discrete element method (DEM) simulations on four types of shared-memory parallel computers: a multicore multiprocessor computer, scalar supercomputer, vector supercomputer, and graphics processing unit. The computational efficiency of a DEM code was found to be drastically improved with our algorithms on all but the scalar supercomputer. Thus, the developed parallel algorithms are useful on shared-memory parallel computers with sufficient memory bandwidth.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

ANALYSIS OF COMMUNICATION PERFORMANCE DEGRADATION OF THE RADIATION TRANSPORT CODE PIDOTS ON HIGH-UTILIZATION, MULTI-USER HPC SYSTEMS

The PIDOTS radiation transport code implements a spatially decomposed Integral Transport Matrix Method (ITMM) solver, which is intended to fully utilize the capabilities of modern, massively parallel high-performance computing (HPC) systems. While the code shows promising results, there was also an unexpected loss of parallel efficiency on the test systems as the number of participating processors grew and the parallelization grain size got finer. We seek to identify and characterize the communication-latency based effects that contribute to this slowdown. Through the creation of a high-level, parameterized representation of the workload of the code, the communication methodology was identified as the likely reason. From there, low-level, InfiniBand communication performance data measured on the Falcon HPC at Idaho National Laboratory (INL) was used to produce a discrete-event model of the communication scheme. The data, as well as the model, show that infrequent, large spikes in send/receive latency drive an unexpectedly substantial increase in projected runtime of a code. Furthermore, these effects are more pronounced under the high-utilization, scattered-workload environment of Falcon. Based on results from the model, it can be shown that a high-communication volume, small message-size regime can easily lead to slowdowns of ~5-10x, when compared to ideal results, as the processor count increases.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2013

YAPPA: a Compiler-Based Parallelization Framework for Irregular Applications on MPSoCs

Modern embedded systems include hundreds of cores. Because of the difficulty in providing a fast, coherent memory architecture, these systems usually rely on non-coherent, non-uniform memory architectures with private memories for each core. However, programming these systems poses significant challenges. The developer must extract large amounts of parallelism, while orchestrating communication among cores to optimize application performance. These issues become even more significant with irregular applications, which present data sets difficult to partition, unpredictable memory accesses, unbalanced control flow and fine grained communication. Hand-optimizing every single aspect is hard and time-consuming, and it often does not lead to the expected performance. There is a growing gap between such complex and highly-parallel architectures and the high level languages used to describe the specification, which were designed for simpler systems and do not consider these new issues. In this paper we introduce YAPPA (Yet Another Parallel Programming Approach), a compilation framework for the automatic parallelization of irregular applications on modern MPSoCs based on LLVM. We start by considering an efficient parallel programming approach for irregular applications on distributed memory systems. We then propose a set of transformations that can reduce the development and optimization effort. The results of our initial prototype confirm the correctness of the proposed approach.

Irregular applications↗
DOE OSTI2011

Reversible Parallel Discrete Event Formulation of a TLM-based Radio Signal Propagation Model

Radio signal strength estimation is essential in many applications, including the design of military radio communications and industrial wireless installations. For scenarios with large or richly- featured geographical volumes, parallel processing is required to meet the memory and computa- tion time demands. Here, we present a scalable and efficient parallel execution of the sequential model for radio signal propagation recently developed by Nutaro et al. Starting with that model, we (a) provide a vector-based reformulation that has significantly lower computational overhead for event handling, (b) develop a parallel decomposition approach that is amenable to reversibility with minimal computational overheads, (c) present a framework for transparently mapping the conservative time-stepped model into an optimistic parallel discrete event execution, (d) present a new reversible method, along with its analysis and implementation, for inverting the vector-based event model to be executed in an optimistic parallel style of execution, and (e) present performance results from implementation on Cray XT platforms. We demonstrate scalability, with the largest runs tested on up to 127,500 cores of a Cray XT5, enabling simulation of larger scenarios and with faster execution than reported before on the radio propagation model. This also represents the first successful demonstration of the ability to efficiently map a conservative time-stepped model to an optimistic discrete-event execution.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2023

High-Level Synthesis of Irregular Applications: A Case Study on Influence Maximization

The Influence Maximization problem is the problem of identifying a small cohort of actors from a broader population that, when initially activated in a diffusion process, are expected to result in a large number of activations in the population. While the problem is known to be NP-hard, several approximation algorithms have been devised by leveraging its submodular structure. While these algorithms are theoretically efficient, they are computationally very expensive in practice. This work advances the current state-of-the-art parallelization scheme for the IMM algorithm by devising the adoption of custom hardware accelerators implemented on FPGAs by leveraging High Level Synthesis from OpenCL. We study the performance of our proposed approach by exploring optimizations tailored at improving the parallel efficiency of the accelerators and highlight their effects and limitations in accelerating complex graph analytic applications. Our experimental evaluation shows that FPGA acceleration can improve the performance of the LT diffusion model up to 1.72x for the entire application and up to 2.90x for its most important kernel with respect to a CPU only parallel execution. The FPGA acceleration of the LT model shows also a 1.54x reduction in energy consumption when compared to a parallel CPU only run.

Neff, Reece W.↗
DOE OSTI2024

Efficient Distributed Sequence Parallelism for Transformer-Based Image Segmentation

We introduce an efficient distributed sequence parallel approach for training transformer-based deep learning image segmentation models. The neural network models are comprised of a combination of a Vision Transformer encoder with a convolutional decoder to provide image segmentation mappings. The utility of the distributed sequence parallel approach is especially useful in cases where the tokenized embedding representation of image data are too large to fit into standard computing hardware memory. To demonstrate the performance and characteristics of our models trained in sequence parallel fashion compared to standard models, we evaluate our approach using a 3D MRI brain tumor segmentation dataset. We show that training with a sequence parallel approach can match standard sequential model training in terms of convergence. Furthermore, we show that our sequence parallel approach has the capability to support training of models that would not be possible on standard computing resources.

Lyngaas, Isaac↗
DOE OSTI2013

Direct Numerical Simulation of Autoignition in a Jet in a Cross-Flow Configuration: ALCF-2 Early Science Program (Technical Report)

Autoignition in turbulent flows is a challenging fundamental problem due to the intricate coupling of different physical and chemical processes extending over multiple flow and chemistry scales. At the same time, the improved understanding and ability to predict autoignition in flows characterized by considerable fluctuations of velocity, composition, and temperature is essential for the development of novel low-emission concepts for power generation. The aim of this project is to study the fundamental aspects of autoignition in a fuel-air mixing device directly applicable to mixing ducts in gas turbines. The NEK5000-based code for low Mach number reactive flows is used to perform very large scale direct numerical simulations of autoignition of a diluted hydrogen jet ejected in a cross-flowing stream of hot turbulent air in a laboratory-scale configuration. We report on our experience running NEK5000 on the new BGQ system at ALCF mira during the early science period (ESP). First of all, the most efficient problem size per MPI-rank is obtained through core-level efficiency metric measured from the target simulation. Furthermore, the most efficient number of ranks is found through strong scaling experiments. Finally, low-level insight into the observed parallel efficiency is enabled through IBM’s HPC Toolkit libraries.

Abdilghanie, A.↗
DOE OSTI2015

MerAligner: A Fully Parallel Sequence Aligner

Aligning a set of query sequences to a set of target sequences is an important task in bioinformatics. In this work we present merAligner, a highly parallel sequence aligner that implements a seed-and-extend algorithm and employs parallelism in all of its components. MerAligner relies on a high performance distributed hash table (seed index) and uses onesided communication capabilities of the Unified Parallel C to facilitate a fine-grained parallelism. We leverage communication optimizations at the construction of the distributed hash table and software caching schemes to reduce communication during the aligning phase. Additionally, merAligner preprocesses the target sequences to extract properties enabling exact sequence matching with minimal communication. Finally, we efficiently parallelize the I/O intensive phases and implement an effective load balancing scheme. Results show that merAligner exhibits efficient scaling up to thousands of cores on a Cray XC30 supercomputer using real human and wheat genome data while significantly outperforming existing parallel alignment tools.

Georganas, E↗
DOE OSTI2016

Partition of unity finite element method for quantum mechanical materials calculations

The current state of the art for large-scale quantum-mechanical simulations is the planewave (PW) pseudopotential method, as implemented in codes such as VASP, ABINIT, and many others. However, since the PW method uses a global Fourier basis, with strictly uniform resolution at all points in space, it suffers from substantial inefficiencies in calculations involving atoms with localized states, such as first-row and transition-metal atoms, and requires significant nonlocal communications, which limit parallel efficiency. Real-space methods such as finite-differences (FD) and finite-elements (FE) have partially addressed both resolution and parallel-communications issues but have been plagued by one key disadvantage relative to PW: excessive number of degrees of freedom (basis functions) needed to achieve the required accuracies. In this paper, we present a real-space partition of unity finite element (PUFE) method to solve the Kohn–Sham equations of density functional theory. In the PUFE method, we build the known atomic physics into the solution process using partition-of-unity enrichment techniques in finite element analysis. The method developed herein is completely general, applicable to metals and insulators alike, and particularly efficient for deep, localized potentials, as occur in calculations at extreme conditions of pressure and temperature. Full self-consistent Kohn–Sham calculations are presented for LiH, involving light atoms, and CeAl, involving heavy atoms with large numbers of atomic-orbital enrichments. We find that the new PUFE approach attains the required accuracies with substantially fewer degrees of freedom, typically by an order of magnitude or more, than the PW method. As a result, we compute the equation of state of LiH and show that the computed lattice constant and bulk modulus are in excellent agreement with reference PW results, while requiring an order of magnitude fewer degrees of freedom to obtain.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2018

A scalable geometric multigrid solver for nonsymmetric elliptic systems with application to variable-density flows

A geometric multigrid algorithm is introduced for solving nonsymmetric linear systems resulting from the discretization of the variable density Navier–Stokes equations on nonuniform structured rectilinear grids and high-Reynolds number flows. The restriction operation is defined such that the resulting system on the coarser grids is symmetric, thereby allowing for the use of efficient smoother algorithms. To achieve an optimal rate of convergence, the sequence of interpolation and restriction operations are determined through a dynamic procedure. A parallel partitioning strategy is introduced to minimize communication while maintaining the load balance between all processors. To test the proposed algorithm, we consider two cases: 1) homogeneous isotropic turbulence discretized on uniform grids and 2) turbulent duct flow discretized on stretched grids. Testing the algorithm on systems with up to a billion unknowns shows that the cost varies linearly with the number of unknowns. This O (N) behavior confirms the robustness of the proposed multigrid method regarding ill-conditioning of large systems characteristic of multiscale high-Reynolds number turbulent flows. The robustness of our method to density variations is established by considering cases where density varies sharply in space by a factor of up to 104, showing its applicability to two-phase flow problems. Strong and weak scalability studies are carried out, employing up to 30,000 processors, to examine the parallel performance of our implementation. Excellent scalability of our solver is shown for a granularity as low as 104 to 105 unknowns per processor. At its tested peak throughput, it solves approximately 4 billion unknowns per second employing over 16,000 processors with a parallel efficiency higher than 50%.

Computer Science↗
DOE OSTI2025

Automatically parallelizing batch inference on deep neural networks using Fiats and Fortran 2023 `do concurrent`

This paper introduces novel programming strategies that leverage features of the Fortran 2023 standard of the International Standards Organization (ISO) to automatically parallelize computations on deep neural networks. The paper focuses on the interplay of object-oriented, parallel, and functional programming paradigms in the Fiats deep learning library. We demonstrate how several infrequently used language features play a role in enabling efficient, parallel execution. Specifically, the ability to explicitly declare that a procedure is pure facilitates inference in the context of the language’s loop-parallelism construct `do concurrent`. Also, explicitly prohibiting the overriding of a parent type’s type-bound procedures eliminates the need for dynamic dispatch in performance-critical code. Finally, this paper uses batch inference calculations on a neural network surrogate for atmospheric aerosol dynamics to demonstrate that LLVM Flang compiler’s automatic parallelization of `do concurrent` achieves roughly the same performance and scalability as achieved by OpenMP compiler directives. We also demonstrate that double-precision inference costs 37–72% longer runtime than default-real precision with most values in the range 57-60%.

Rouson, Damian↗