Search NASA⌕ Search

SEARCH · Search NASA

Results for “data parallel algorithms”

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 OSTI2024

AEOLUS: Advances in Experimental Design, Optimal Control, and Learning for Uncertain Complex Systems

Sustained advances in the mathematics of modeling and simulation have resulted in the capability today for routine simulation of a number of large scale complex DOE-relevant systems. As remarkable as this capability for solving the so-called forward problem is, it is typically only the first step-an inner loop within an outer loop that explores the simulation model's parameter space and decision space to characterize uncertainty in the model's predictions, learn unknown model parameters from data, design the most informative experiments, determine optimal control strategies, and create optimal designs. Broadly, what unifies all of these outer loop problems is that they are, in one form or another, optimization problems over parameter/control/design space that are constrained by complex uncertain models. To fully realize the power of scientific simulation as a basis for scientific discovery, technological innovation, and rational decision-making, it is imperative to move beyond simulation to tackle the outer loop of optimization for learning from data, experimental design, and control with complex uncertain models. When the models under consideration are large-scale and complex, and when the optimization variable and uncertain parameter spaces are high (or infinite) dimensional, this constitutes a grand challenge of the highest order, and is intractable with conventional methods. To overcome these challenges, the AEOLUS Center was established to develop a unified mathematical, computational, and statistical framework for (1) Learning predictive models from complex data via Bayesian inference and optimization, and (2) Optimizing experiments, processes, and designs using the resulting uncertain models. These problems are intractable with conventional methods, for several reasons: (1) The simulation problems that govern the inner loops of the optimization problems are expensive to execute (due to severe nonlinearity, heterogeneity, multiphysics/multiscale coupling); (2) The optimization variable and uncertain parameter spaces are high dimensional, often stemming from discretizations of infinite dimensional fields such as initial conditions, sources, or material properties. We argue that the key to overcoming these challenges is to develop new mathematical, computational, and statistical methods that exploit the structure of the Bayesian inference and optimization problems mediated by their underlying complex uncertain models. This structure includes the regularity, sparsity, geometry, low intrinsic dimensionality, and multifidelity nature of the maps from uncertain parameter/optimization variable spaces to the specific objectives targeted: Bayesian inference, optimal experimental design, and optimal control design. Black box methods developed as generic tools are incapable of exploiting this structure. To be successful, we must create, integrate, and cross-fertilize ideas across multiple areas of applied math--including approximation theory, Bayesian inference, data science, experimental design, information theory, machine learning, model reduction, optimal control theory, parallel algorithms, PDE-constrained optimization, randomized algorithms, stochastic optimization, and uncertainty quantification--all while exploiting the structure of the problems at hand. With this goal in mind, we have marshaled a team of leading authorities in these areas. While the methods we develop will be broadly applicable across a wide spectrum of DOE problems in which experiments inform models and the systems those models describe must be optimized under uncertainty, we have chosen a specific area, advanced manufacturing and materials, to drive our work. AMM is characterized by complex models across multiple scales, and is a rich source of challenging problems in inference, experimental design, and optimal control, requiring multifaceted and integrated advances in applied mathematics. As such, AMM serves as an excellent vehicle to motivate and demonstrate the advances in applied mathematics developed by our center.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Massively parallel modeling and inversion of electrical resistivity tomography data using PFLOTRAN

Abstract. Electrical resistivity tomography (ERT) is a broadly accepted geophysical method for subsurface investigations. Interpretation of field ERT data usually requires the application of computationally intensive forward modeling and inversion algorithms. For large-scale ERT data, the efficiency of these algorithms depends on the robustness, accuracy, and scalability on high-performance computing resources. In this regard, we present a robust and highly scalable implementation of forward modeling and inversion algorithms for ERT data. The implementation is publicly available and developed within the framework of PFLOTRAN, an open-source, state-of-the-art massively parallel subsurface flow and transport simulation code. The forward modeling is based on a finite-volume discretization of the governing differential equations, and the inversion uses a Gauss–Newton optimization scheme. To evaluate the accuracy of the forward modeling, two examples are first presented by considering layered (1D) and 3D earth conductivity models. The computed numerical results show good agreement with the analytical solutions for the layered earth model and results from a well-established code for the 3D model. Inversion of ERT data, simulated for a 3D model, is then performed to demonstrate the inversion capability by recovering the conductivity of the model. To demonstrate the parallel performance of PFLOTRAN's ERT process model and inversion capabilities, large-scale scalability tests are performed by using up to 131 072 processes on a leadership class supercomputer. These tests are performed for the two most computationally intensive steps of the ERT inversion: forward modeling and Jacobian computation. For the forward modeling, we consider models with up to 122 ×106 degrees of freedom (DOFs) in the resulting system of linear equations and demonstrate that the code exhibits almost linear scalability on up to 10 000 DOFs per process. On the other hand, the code shows superlinear scalability for the Jacobian computation, mainly because all computations are fairly evenly distributed over each process with no parallel communication.

58 GEOSCIENCES↗
DOE OSTI2021

Improving I/O Performance for Exascale Applications through Online Data Layout Reorganization

The applications being developed within the U.S. Exascale Computing Project (ECP) to run on imminent Exascale computers will generate scientific results with unprecedented fidelity and record turn-around time. Many of these codes are based on particle-mesh methods and use advanced algorithms, especially dynamic load-balancing and mesh-refinement, to achieve high performance on Exascale machines. Yet, as such algorithms improve parallel application efficiency, they raise new challenges for I/O logic due to their irregular and dynamic data distributions. Thus, while the enormous data rates of Exascale simulations already challenge existing file system write strategies, the need for efficient read and processing of generated data introduces additional constraints on the data layout strategies that can be used when writing data to secondary storage. We review these I/O challenges and introduce two online data layout reorganization approaches for achieving good tradeoffs between read and write performance. We demonstrate the benefits of using these two approaches for the ECP particle-in-cell simulation WarpX, which serves as a motif for a large class of important Exascale applications. Here, we show that by understanding application I/O patterns and carefully designing data layouts we can increase read performance by more than 80 percent.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Understanding performance variability in standard and pipelined parallel Krylov solvers

In this work, we collect data from runs of Krylov subspace methods and pipelined Krylov algorithms in an effort to understand and model the impact of machine noise and other sources of variability on performance. We find large variability of Krylov iterations between compute nodes for standard methods that is reduced in pipelined algorithms, directly supporting conjecture, as well as large variation between statistical distributions of runtimes across iterations. Based on these results, we improve upon a previously introduced nondeterministic performance model by allowing iterations to fluctuate over time. We present our data from runs of various Krylov algorithms across multiple platforms as well as our updated non-stationary model that provides good agreement with observations. We also suggest how it can be used as a predictive tool.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

In situ multi-tier auto-ignition detection applied to dual-fuel combustion simulations

Here we use an anomaly detection methodology that is centered on analyzing fourth-order joint moments (co-kurtosis), particularly focusing on its application in auto-ignition of combustion problems with large numbers of species. Unsupervised anomaly detection is challenging to generalize across problem types and domains. A recent technique, centered on analyzing information in the fourth-order joint moment co-kurtosis, has shown promise, especially for high-dimensional scientific data. In this work we present developments to the co-kurtosis based anomaly detection method needed to make it effective and scalable for large-scale distributed scientific data, such as those generated by massively parallel simulations. An in situ co-kurtosis algorithm is employed as the anomaly detection method for identifying ignition kernels in simulations of turbulent combustion. Here, we extend an existing methodology which identifies regions of the domain where anomalies are present, and add another tier of anomaly detection where the individual samples contributing to the anomaly are identified. We apply this algorithm on-the-fly to a variety of turbulent reacting flow problems and compare it to the widely used (but significantly more expensive) chemical explosive mode analysis (CEMA). We demonstrate the ability of the method to detect and identify the onset of low and high temperature ignition which can be used for computational steering, as chemical and combustion anomalies occur intermittently at spatio-temporal locations unknown a priori. Finally, we apply our lightweight in situ algorithm to an exascale high-fidelity simulation with a total of 2.4 Trillion degrees of freedom, performed using an adaptive mesh refinement solver. Furthermore, through a scalability analysis, we show that the relative computational cost of this in-situ anomaly detection algorithm compared to an iteration of the reacting flow solver is negligible.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Data Locality Enhancement of Dynamic Simulations for Exascale Computing (Final Report)

The development of modern processors exhibits two trends that complicate the optimizations of modern software. The first is the increasing sensitivity of processors' throughput to irregularities in computation. With more processors produced through a massive integration of simple cores, future systems will increasingly favor regular data-level parallel computations, but deviate from the needs of applications with complex patterns. Some evidences are already shown on Graphic Processing Units (GPU): Irregular data accesses (e.g., indirect references A[D[i]]) and conditional branches are limiting many GPU applications' performance at a level an order of magnitude lower than the peak of GPU. The second hardware trend is the growing gap between memory bandwidth and the aggregate speed—that is, the sum of all cores' computing power—of a Chip Multiprocessor (CMP). Despite the capped growth of the peak CPU speed, the aggregate speed of a CMP keeps increasing as more cores get into a single chip. It is expected that by 2018, node concurrency in an exascale system will increase by hundreds of times, whereas, memory bandwidth will expand by only 10 to 20 times. Consequently, data movement and storage is expected to consume more than 70% of the total system power. Bridging this gap is difficult; the complexities of modern CMP memory hierarchy make it even harder: Data cache becomes shared among computing units, and the sharing is often non-uniform—whether two computing units share a cache depends on their proximity and the level of the cache. On the recent IBM Power7 architecture, for instance, four hardware contexts (or SMT threads) in a core share the entire memory hierarchy, all cores in one chip share an on-chip L3 cache, and cores across chips share L3 and main memory through off-chip connections. These two trends complicate the translation of computing power into performance, especially for a program with either intensive data accesses or complex patterns in data accesses or control flow paths. Unfortunately, both attributes present and will persist in a class of important applications. For instance, many scientific simulations deal with a large volume of data. And meanwhile, as most real-world processes are non-uniform and evolving (e.g., the evolution of a galaxy or the process of a drug injection), both the computations and data accesses of these programs tend to be irregular and dynamically changing. Currently, the lack of support to these applications on modern CMP severely limits their performance. On GPU, as our recent study shows and other studies echo, performance enhancement of a factor of integers is possible when memory accesses or control flows are streamlined for a set of GPU applications. On multicore CPU, our studies show that traditional locality enhancement, for being oblivious to the new features of multicore memory hierarchy, may even cause large slowdown to data-intensive dynamic applications. The severity of the issues is expected to worsen as the two hardware trends continue. Some recent studies try to match software with the trends, but in a limited scope or manner. For irregularities on GPU, most studies focus on irregularities analyzable through static analysis (e.g., data accesses in regular loops). Dynamic irregularities are harder to address because the needed analysis and transformations typically have to happen at run time. Some other research resorts to hardware extensions, an actual adoption of which is unclear for the entailed space cost and complexity. For data locality, recent years have seen some exploitations of the new memory hierarchy on multicore for performance, but most of them are on process or thread scheduling, rather than program transformations. Our recent study reveals that program-level transformations may magnify the scheduling benefits by a factor of seven, concluding that program-level transformation should play a central role for data locality enhancement on modern CMP. But research in this direction has been sparse, and most have focused on data layout or cache performance modeling, rather than program transformations to match with the new memory hierarchy features. Overall, it is still an open question how to bridge the gap between dynamic computations and the two prominent properties of modern processors. The goal of this project is to develop a set of techniques and software tools to enhance the matching between memory accesses in dynamic simulations and the prominent features of modern and future CMP, alleviating the memory performance issues for petascale and exascale computing. This report summarizes the discoveries and products produced throughout this project. It includes free launch, a new software approach to overcoming the shortcomings of both methods; coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution; algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Supporting dynamic parallelism is important for GPU to benefit a broad range of applications. There are currently two fundamental ways for programs to exploit dynamic parallelism on GPU: a software-based approach with software-managed worklists, and a hardware-based approach through dynamic subkernel launches. Neither is satisfactory. The former is complicated to program and is often subject to some load imbalance; the latter suffers large runtime overhead. In this work, we propose free launch, a new software approach to overcoming the shortcomings of both methods. It allows programmers to use subkernel launches to express dynamic parallelism. It employs a novel compiler-based code transformation named subkernel launch removal to replace the subkernel launches with the reuse of parent threads. Coupled with an adaptive task assignment mechanism, the transformation reassigns the tasks in the subkernels to the parent threads with a good load balance. The technique requires no hardware extensions, immediately deployable on existing GPUs. It keeps the programming convenience of the subkernel launch-based approach while avoiding its large runtime overhead. Meanwhile, its superior load balancing makes it outperform manual worklist-based techniques by 3X on average. The work was published at the 48th Annual IEEE/ACM International Symposium on Microarchitecture (Micro'2015). A Graphic Processing Unit (GPU) system is typically equipped with many types of memory (e.g., global, constant, texture, shared, cache). Data placement determines what data are placed on which type of memory, essential for GPU memory performance. Prior optimizations of data placement always require a single view of a data object on memory, which limits the optimization effectiveness. In this work, we propose coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution. We demonstrate that under certain conditions, the multiple views can remain incoherent while facilitating enhanced data placement. We present a theorem and some compiler support to ensure the soundness of the usage of coherence-free multiview. We further develop reference-discerning data placement, a new way to enhance data placements on GPU. It enables more flexible data placements by using coherence-free multiview to leverage the slack in coherence requirement of some GPU programs. Experiments on three types of GPU systems show that, with less than 200KB space cost, the new data placement technique can provide a 1.6X average (up to 4.27X) speedup. In addition, we have examined algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Computing distances among data points is an essential part of many important algorithms in data analytics, graph analysis, and other domains. In each of these domains, developers have spent significant manual effort optimizing algorithms, often through novel applications of the triangle equality, in order to minimize the number of distance computations in the algorithms. In this work, we observe that many algorithms across these domains can be generalized as an instance of a generic distance-related abstraction. Based on this abstraction, we derive seven principles for correctly applying the triangular inequality to optimize distance-related algorithms. Guided by the findings, we develop Triangular OPtimizer (TOP), the first software framework that is able to automatically produce optimized algorithms that either matches or outperforms manually designed algorithms for solving distance-related problems. TOP achieves up to 237x speedups and 2.5X on average. The work has been published at the 32nd International Conference on Machine Learning (ICML'15) and the 41st International Conference on Very Large Data Bases (VLDB'15).

42 ENGINEERING↗
DOE OSTI2020

Algorithms for Efficient Reproducible Floating Point Summation

We define “reproducibility” as getting bitwise identical results from multiple runs of the same program, perhaps with different hardware resources or other changes that should not affect the answer. Many users depend on reproducibility for debugging or correctness. However, dynamic scheduling of parallel computing resources, combined with nonassociative floating point addition, makes reproducibility challenging even for summation, or operations like the BLAS. We describe a “reproducible accumulator” data structure (the “binned number”) and associated algorithms to reproducibly sum binary floating point numbers, independent of summation order. We use a subset of the IEEE Floating Point Standard 754-2008 and bitwise operations on the standard representations in memory. Our approach requires only one read-only pass over the data, and one reduction in parallel, using a 6-word reproducible accumulator (more words can be used for higher accuracy), enabling standard tiling optimization techniques. Summing n words with a 6-word reproducible accumulator requires approximately 9 n floating point operations (arithmetic, comparison, and absolute value) and approximately 3 n bitwise operations. The final error bound with a 6-word reproducible accumulator and our default settings can be up to 2 29 times smaller than the error bound for conventional (recursive) summation on ill-conditioned double-precision inputs.

Computer Science↗
DOE OSTI2025

New Results on Communication- and Memory-Aware Load Balancing Model and Algorithms

While load balancing in distributed-memory computing has been well-studied, we present an innovative approach to this problem: a unified, reduced-order model that combines three key components to describe “work” in a distributed system: computation, communication, and memory. Our model enables an optimizer to explore complex tradeoffs in task placement, such as augmented parallelism, at the expense of data replication increasing memory usage. We propose a fully distributed, heuristic-based load balancing optimization algorithm, and demonstrate that it quickly finds close-to-optimal solutions. We formalize the complex optimization problem as a mixed-integer linear program, and compare it to our strategy. Finally, we show that when applied to an electromagnetics code, our approach obtains up to 2.3x speedups for the imbalanced execution.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

PASTA: A Parallel Sparse Tensor Algorithm Benchmark Suite

Tensor methods have gained increasingly attention from various applications, including machine learning, quantum chemistry, healthcare analytics, social network analysis, data mining, and signal processing, to name a few. Sparse tensors and their algorithms become critical to further improve the performance of these methods and enhance the interpretability of their output. This work presents a sparse tensor algorithm benchmark suite (PASTA) for single- and multi-core CPUs. To the best of our knowledge, this is the first benchmark suite for sparse tensor world. PASTA targets on: 1) helping application users to evaluate different computer systems using its representative computational workloads; 2) providing insights to better utilize existed computer architecture and systems and inspiration for the future design. This benchmark suite will be publicly released.

HPC, sparse tensor, benchmark↗
DOE OSTI2023

Distributed Data-Driven Optimization for Voltage Regulation in Distribution Systems

Here, this paper proposes a distributed data-driven optimization framework for voltage regulation in distribution systems. The recursive kernel regression and alternating direction method of multipliers (ADMM) are selected to cover the system learning and distributed optimization tasks. The proposed distributed data-driven framework is capable of having a rapid response to system or load changes while considering the operation optimality. Besides, the distributed algorithm parallels the computation tasks and reduces the computational expense of a single agent. To validate the performance of the proposed method, a hypothetical 7-Bus system and the IEEE 123-Bus system are selected to show the effectiveness of the proposed data-driven framework. According to the numerical study results, the proposed method offers great flexibility for selecting customized kernel models for different regions and can effectively improve the system voltage profile in a distributed manner.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

PeleLMeX: an AMR Low Mach Number Reactive Flow Simulation Code without level sub-cycling

PeleLMeX simulates chemically reacting low Mach number flows with block-structured adaptive mesh refinement (AMR). The code is built upon the AMReX library, which provides the underlying data structures and tools to manage and operate on them across massively parallel computing architectures. PeleLMeX algorithmic features are inherited from its predecessor PeleLM but key improvements allow representation of more complex physical processes. Together with its compressible flow counterpart PeleC, the thermo-chemistry library PelePhysics and the multi-physics library PeleMP, it forms the Pele suite of open-source reactive flow simulation codes.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

PyFLEXTRKR: a flexible feature tracking Python software for convective cloud analysis

Abstract. This paper describes the new open-source framework PyFLEXTRKR (Python FLEXible object TRacKeR), a flexible atmospheric feature tracking software package with specific capabilities to track convective clouds from a variety of observations and model simulations. This software can track any atmospheric 2D objects and handle merging and splitting explicitly. The package has a collection of multi-object identification algorithms, scalable parallelization options, and has been optimized for large datasets including global high-resolution data. We demonstrate applications of PyFLEXTRKR on tracking individual deep convective cells and mesoscale convective systems from observations and model simulations ranging from large-eddy resolving (∼100s m) to mesoscale (∼10s km) resolutions. Visualization, post-processing, and statistical analysis tools are included in the package. New Lagrangian analyses of convective clouds produced by PyFLEXTRKR applicable to a wide range of datasets and scales facilitate advanced model evaluation and development efforts as well as scientific discovery.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Efficient exascale discretizations: High-order finite element methods

Efficient exploitation of exascale architectures requires rethinking of the numerical algorithms used in many large-scale applications. These architectures favor algorithms that expose ultra fine-grain parallelism and maximize the ratio of floating point operations to energy intensive data movement. One of the few viable approaches to achieve high efficiency in the area of PDE discretizations on unstructured grids is to use matrix-free/partially assembled high-order finite element methods, since these methods can increase the accuracy and/or lower the computational time due to reduced data motion. In this paper we provide an overview of the research and development activities in the Center for Efficient Exascale Discretizations (CEED), a co-design center in the Exascale Computing Project that is focused on the development of next-generation discretization software and algorithms to enable a wide range of finite element applications to run efficiently on future hardware. CEED is a research partnership involving more than 30 computational scientists from two US national labs and five universities, including members of the Nek5000, MFEM, MAGMA and PETSc projects. We discuss the CEED co-design activities based on targeted benchmarks, miniapps and discretization libraries and our work on performance optimizations for large-scale GPU architectures. We also provide a broad overview of research and development activities in areas such as unstructured adaptive mesh refinement algorithms, matrix-free linear solvers, high-order data visualization, and list examples of collaborations with several ECP and external applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Parallel physics-informed neural networks via domain decomposition

Here we develop a distributed framework for the physics-informed neural networks (PINNs) based on two recent extensions, namely conservative PINNs (cPINNs) and extended PINNs (XPINNs), which employ domain decomposition in space and in time-space, respectively. This domain decomposition endows cPINNs and XPINNs with several advantages over the vanilla PINNs, such as parallelization capacity, large representation capacity, efficient hyperparameter tuning, and is particularly effective for multi-scale and multi-physics problems. Here, we present a parallel algorithm for cPINNs and XPINNs constructed with a hybrid programming model described by MPI + X, where X ∈ {CPUs, GPUs}. The main advantage of cPINN and XPINN over the more classical data and model parallel approaches is the flexibility of optimizing all hyperparameters of each neural network separately in each subdomain. We compare the performance of distributed cPINNs and XPINNs for various forward problems, using both weak and strong scalings. Our results indicate that for space domain decomposition, cPINNs are more efficient in terms of communication cost but XPINNs provide greater flexibility as they can also handle time-domain decomposition for any differential equations, and can deal with any arbitrarily shaped complex subdomains. To this end, we also present an application of the parallel XPINN method for solving an inverse diffusion problem with variable conductivity on the United States map, using ten regions as subdomains.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

High-performance computing in water resources hydrodynamics

In this work, we present a vision of future water resources hydrodynamics codes that can fully utilize the strengths of modern high-performance computing. The advances to computing power, formerly driven by the improvement of central processing unit processors, now focus on parallel computing and, in particular, the use of graphics processing units (GPUs). However, this shift to a parallel framework requires refactoring the code to make efficient use of the data as well as changing even the nature of the algorithm that solves the system of equations. These concepts along with other features such as the precision for the computations, dry regions management, and input/output data are analyzed in this paper. A 2D multi-GPU flood code applied to a large-scale test case is used to corroborate our statements and ascertain the new challenges for the next-generation parallel water resources codes.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2019

Tail queues: A multi-threaded matching architecture

As we approach exascale, computational parallelism will have to drastically increase in order to meet throughput targets. Many–core architectures have exacerbated this problem by trading reduced clock speeds, core complexity, and computation throughput for increasing parallelism. This presents two major challenges for communication libraries such as MPI: the library must leverage the performance advantages of thread level parallelism and avoid the scalability problems associated with increasing the number of processes to that scale. Hybrid programming models, such as MPI+X, have been proposed to address these challenges. MPI THREAD MULTIPLE is MPI's thread safe mode. While there has been work to optimize it, it largely remains non–performant in most implementations. While current applications avoid MPI multithreading due to performance concerns, it is expected to be utilized in future applications. One of the major synchronous data structures required by MPI is the matching engine. In this paper, we present a parallel matching algorithm that can improve MPI matching for multithreaded applications. We then perform a feasibility study to demonstrate the performance benefit of the technique.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

GentenMPI: Distributed Memory Sparse Tensor Decomposition

GentenMPl is a toolkit of sparse canonical polyadic (CP) tensor decomposition algorithms that is designed to run effectively on distributed-memory high-performance computers. Its use of distributed-memory parallelism enables it to efficiently decompose tensors that are too large for a single compute node's memory. GentenMPl leverages Sandia's decades-long investment in the Trilinos solver framework for much of its parallel-computation capability. Trilinos contains numerical algorithms and linear algebra classes that have been optimized for parallel simulation of complex physical phenomena. This work applies these tools to the data science problem of sparse tensor decomposition. In this report, we describe the use of Trilinos in GentenMPl, extensions needed for sparse tensor decomposition, and implementations of the CP-ALS (CP via alternating least squares) and GCP-SGD (generalized CP via stochastic gradient descent) sparse tensor decomposition algorithms. We show that GentenMPl can decompose sparse tensors of extreme size, e.g., a 12.6-terabyte tensor on 8192 computer cores. We demonstrate that the Trilinos backbone provides good strong and weak scaling of the tensor decomposition algorithms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Performance Analysis of Traditional and Data-Parallel Primitive Implementations of Visualization and Analysis Kernels

Measurements of absolute runtime are useful as a summary of performance when studying parallel visualization and analysis methods on computational platforms of increasing concurrency and complexity. We can obtain even more insights by measuring and examining more detailed measures from hardware performance counters, such as the number of instructions executed by an algorithm implemented in a particular way, the amount of data moved to/from memory, memory hierarchy utilization levels via cache hit/miss ratios, and so forth. This work focuses on performance analysis on modern multi-core platforms of three different visualization and analysis kernels that are implemented in different ways: one is "traditional", using combinations of C++ and VTK, and the other uses a data-parallel approach using VTK-m. Our performance study consists of measurement and reporting of several different hardware performance counters on two different multi-core CPU platforms. The results reveal interesting performance differences between these two different approaches for implementing these kernels, results that would not be apparent using runtime as the only metric.

97 MATHEMATICS AND COMPUTING↗