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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

Improvements to a Batch Pentadiagonal Solver on NVIDIA GPUs

This poster presents the recent work in OVERFLOW to port the batched pentadiagonal solver to NVIDIA GPUs. There are five pentadiagonal systems for each pencil in the grid but three of these systems share the same LHS. Our first simple approach for porting the pentadiagonal solver to the GPUs was to take advantage of the shared LHS by assigning three threads to the three LHS of each pencil. We demonstrated that this custom solver was 92% faster than the NVIDIA batched pentadiagonal library implementation on a V100 GPU due to the lower memory bandwidth requirements. The second approach treated each pentadiagonal system as a 2x2 block tridiagonal system and used a variant of the parallel cyclic reduction algorithm to solve the problem. One benefit of this approach is that it does not require interleaving the data between each system. We demonstrated that this algorithm is 2.18x faster than the NVIDIA library implementation for the same amount of work. If we take advantage of our shared LHS, this approach is 2.58x faster than the library implementation on a V100 GPU.

GPU Programming↗
NASA NTRS1998

I/O Parallelization for the Goddard Earth Observing System Data Assimilation System (GEOS DAS)

The National Aeronautics and Space Administration (NASA) Data Assimilation Office (DAO) at the Goddard Space Flight Center (GSFC) has developed the GEOS DAS, a data assimilation system that provides production support for NASA missions and will support NASA's Earth Observing System (EOS) in the coming years. The GEOS DAS will be used to provide background fields of meteorological quantities to EOS satellite instrument teams for use in their data algorithms as well as providing assimilated data sets for climate studies on decadal time scales. The DAO has been involved in prototyping parallel implementations of the GEOS DAS for a number of years and is now embarking on an effort to convert the production version from shared-memory parallelism to distributed-memory parallelism using the portable Message-Passing Interface (MPI). The GEOS DAS consists of two main components, an atmospheric General Circulation Model (GCM) and a Physical-space Statistical Analysis System (PSAS). The GCM operates on data that are stored on a regular grid while PSAS works with observational data that are scattered irregularly throughout the atmosphere. As a result, the two components have different data decompositions. The GCM is decomposed horizontally as a checkerboard with all vertical levels of each box existing on the same processing element(PE). The dynamical core of the GCM can also operate on a rotated grid, which requires communication-intensive grid transformations during GCM integration. PSAS groups observations on PEs in a more irregular and dynamic fashion.

Lucchesi, Rob↗
NASA NTRS1989

Parallel implementation and evaluation of motion estimation system algorithms on a distributed memory multiprocessor using knowledge based mappings

Several techniques to perform static and dynamic load balancing techniques for vision systems are presented. These techniques are novel in the sense that they capture the computational requirements of a task by examining the data when it is produced. Furthermore, they can be applied to many vision systems because many algorithms in different systems are either the same, or have similar computational characteristics. These techniques are evaluated by applying them on a parallel implementation of the algorithms in a motion estimation system on a hypercube multiprocessor system. The motion estimation system consists of the following steps: (1) extraction of features; (2) stereo match of images in one time instant; (3) time match of images from different time instants; (4) stereo match to compute final unambiguous points; and (5) computation of motion parameters. It is shown that the performance gains when these data decomposition and load balancing techniques are used are significant and the overhead of using these techniques is minimal.

Choudhary, Alok Nidhi↗
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↗
NASA NTRS1975

Evaluation of ERTS multispectral signatures in relation to ground control signatures using a nested-sampling approach

The author has identified the following significant results. Ground measured spectral signatures of wavelength bands matching ERTS MSS were collected using a radiometer at several Californian and Nevadan sites, and directly compared with similar data from ERTS CCTs. The comparison was tested at the highest possible spatial resolution for ERTS, using deconvoluted MSS data, and contrasted with that of ground measured spectra, originally from 1 meter squares. In the mobile traverses of the grassland sites, these one meter fields of view were integrated into eighty meter transects along the five km track across four major rock/soil types. Suitable software was developed to read the MSS CCT tapes, to shadeprint individual bands with user-determined greyscale stretching. Four new algorithms for unsupervised and supervised, normalized and unnormalized clustering were developed, into a program termed STANSORT. Parallel software allowed the field data to be calibrated, and by using concurrently continuously collected, upward- and downward-viewing, 4 band radiometers, bidirectional reflectances could be calculated.

Lyon, R. J. P.↗
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↗
NASA NTRS1994

An implementation of a tree code on a SIMD, parallel computer

We describe a fast tree algorithm for gravitational N-body simulation on SIMD parallel computers. The tree construction uses fast, parallel sorts. The sorted lists are recursively divided along their x, y and z coordinates. This data structure is a completely balanced tree (i.e., each particle is paired with exactly one other particle) and maintains good spatial locality. An implementation of this tree-building algorithm on a 16k processor Maspar MP-1 performs well and constitutes only a small fraction (approximately 15%) of the entire cycle of finding the accelerations. Each node in the tree is treated as a monopole. The tree search and the summation of accelerations also perform well. During the tree search, node data that is needed from another processor is simply fetched. Roughly 55% of the tree search time is spent in communications between processors. We apply the code to two problems of astrophysical interest. The first is a simulation of the close passage of two gravitationally, interacting, disk galaxies using 65,636 particles. We also simulate the formation of structure in an expanding, model universe using 1,048,576 particles. Our code attains speeds comparable to one head of a Cray Y-MP, so single instruction, multiple data (SIMD) type computers can be used for these simulations. The cost/performance ratio for SIMD machines like the Maspar MP-1 make them an extremely attractive alternative to either vector processors or large multiple instruction, multiple data (MIMD) type parallel computers. With further optimizations (e.g., more careful load balancing), speeds in excess of today's vector processing computers should be possible.

Olson, Kevin M.↗
NASA NTRS1986

Algorithms and programming tools for image processing on the MPP, part 2

A number of algorithms were developed for image warping and pyramid image filtering. Techniques were investigated for the parallel processing of a large number of independent irregular shaped regions on the MPP. In addition some utilities for dealing with very long vectors and for sorting were developed. Documentation pages for the algorithms which are available for distribution are given. The performance of the MPP for a number of basic data manipulations was determined. From these results it is possible to predict the efficiency of the MPP for a number of algorithms and applications. The Parallel Pascal development system, which is a portable programming environment for the MPP, was improved and better documentation including a tutorial was written. This environment allows programs for the MPP to be developed on any conventional computer system; it consists of a set of system programs and a library of general purpose Parallel Pascal functions. The algorithms were tested on the MPP and a presentation on the development system was made to the MPP users group. The UNIX version of the Parallel Pascal System was distributed to a number of new sites.

Reeves, Anthony P.↗
NASA NTRS2002

Field-Programmable Gate Array Computer in Structural Analysis: An Initial Exploration

This paper reports on an initial assessment of using a Field-Programmable Gate Array (FPGA) computational device as a new tool for solving structural mechanics problems. A FPGA is an assemblage of binary gates arranged in logical blocks that are interconnected via software in a manner dependent on the algorithm being implemented and can be reprogrammed thousands of times per second. In effect, this creates a computer specialized for the problem that automatically exploits all the potential for parallel computing intrinsic in an algorithm. This inherent parallelism is the most important feature of the FPGA computational environment. It is therefore important that if a problem offers a choice of different solution algorithms, an algorithm of a higher degree of inherent parallelism should be selected. It is found that in structural analysis, an 'analog computer' style of programming, which solves problems by direct simulation of the terms in the governing differential equations, yields a more favorable solution algorithm than current solution methods. This style of programming is facilitated by a 'drag-and-drop' graphic programming language that is supplied with the particular type of FPGA computer reported in this paper. Simple examples in structural dynamics and statics illustrate the solution approach used. The FPGA system also allows linear scalability in computing capability. As the problem grows, the number of FPGA chips can be increased with no loss of computing efficiency due to data flow or algorithmic latency that occurs when a single problem is distributed among many conventional processors that operate in parallel. This initial assessment finds the FPGA hardware and software to be in their infancy in regard to the user conveniences; however, they have enormous potential for shrinking the elapsed time of structural analysis solutions if programmed with algorithms that exhibit inherent parallelism and linear scalability. This potential warrants further development of FPGA-tailored algorithms for structural analysis.

Singleterry, Robert C., Jr.↗
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↗
DOE OSTI2019

A communication-avoiding 3D sparse triangular solver

We present a novel distributed memory algorithm to improve the strong scalability of the solution of a sparse triangular system. This operation appears in the solve phase of direct methods for solving general sparse linear systems, Ax = b. Our 3D sparse triangular solver employs several techniques, including a 3D MPI process grid, elimination tree parallelism, and data replication, all of which reduce the per-process communication when combined. We present analytical models to understand the communication cost of our algorithm and show that our 3D sparse triangular solver can reduce the per-process communication volume asymptotically by a factor of O(n1/4) and O(n1/6) for problems arising from the finite element discretizations of 2D "planar" and 3D "non-planar" PDEs, respectively. We implement our algorithm for use in SuperLU_DIST3D, using a hybrid MPI+OpenMP programming model. Our 3D triangular solve algorithm, when run on 12k cores of Cray XC30, outperforms the current state-of-the-art 2D algorithm by 7.2x for planar and 2.7x for the non-planar sparse matrices, respectively.

Sao, Piyush↗
NASA NTRS1987

A dual-processor multi-frequency implementation of the FINDS algorithm

This report presents a parallel processing implementation of the FINDS (Fault Inferring Nonlinear Detection System) algorithm on a dual processor configured target flight computer. First, a filter initialization scheme is presented which allows the no-fail filter (NFF) states to be initialized using the first iteration of the flight data. A modified failure isolation strategy, compatible with the new failure detection strategy reported earlier, is discussed and the performance of the new FDI algorithm is analyzed using flight recorded data from the NASA ATOPS B-737 aircraft in a Microwave Landing System (MLS) environment. The results show that low level MLS, IMU, and IAS sensor failures are detected and isolated instantaneously, while accelerometer and rate gyro failures continue to take comparatively longer to detect and isolate. The parallel implementation is accomplished by partitioning the FINDS algorithm into two parts: one based on the translational dynamics and the other based on the rotational kinematics. Finally, a multi-rate implementation of the algorithm is presented yielding significantly low execution times with acceptable estimation and FDI performance.

Godiwala, Pankaj M.↗
DOE OSTI2018

167-PFlops Deep Learning for Electron Microscopy: From Learning Physics to Atomic Manipulation

An artificial intelligence system called MENNDL, which used 25,200 Nvidia Volta GPUs on Oak Ridge National Laboratory’s Summit machine, automatically designed an optimal deep learning network in order to extract structural information from raw atomic-resolution microscopy data. In a few hours, MENNDL creates and evaluates millions of networks using a scalable, parallel, asynchronous genetic algorithm augmented with a support vector machine to automatically find a superior deep learning network topology and hyper-parameter set than a human expert can find in months. For the application of electron microscopy, the system furthers the goal of improving our understanding of the electron-beam-matter interactions and real-time image-based feedback, which enables a huge step beyond human capacity toward nanofabricating materials automatically. Finally, MENNDL has been scaled to the 4,200 available nodes of Summit achieving a measured 152.5 PFlops, with an estimated sustained performance of 167 PFlops when the entire machine is available.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2000

"Genetically Engineered" Nanoelectronics

The quantum mechanical functionality of nanoelectronic devices such as resonant tunneling diodes (RTDs), quantum well infrared-photodetectors (QWIPs), quantum well lasers, and heterostructure field effect transistors (HFETs) is enabled by material variations on an atomic scale. The design and optimization of such devices requires a fundamental understanding of electron transport in such dimensions. The Nanoelectronic Modeling Tool (NEMO) is a general-purpose quantum device design and analysis tool based on a fundamental non-equilibrium electron transport theory. NEW was combined with a parallelized genetic algorithm package (PGAPACK) to evolve structural and material parameters to match a desired set of experimental data. A numerical experiment that evolves structural variations such as layer widths and doping concentrations is performed to analyze an experimental current voltage characteristic. The genetic algorithm is found to drive the NEMO simulation parameters close to the experimentally prescribed layer thicknesses and doping profiles. With such a quantitative agreement between theory and experiment design synthesis can be performed.

Klimeck, Gerhard↗
DOE OSTI2018

167-PFlops deep learning for electron microscopy: from learning physics to atomic manipulation

An artificial intelligence system called MENNDL, which used 25,200 NVIDIA Volta GPUs on Oak Ridge National Laboratory's Summit machine, automatically designed an optimal deep learning network in order to extract structural information from raw atomic-resolution microscopy data. In a few hours, MENNDL creates and evaluates millions of networks using a scalable, parallel, asynchronous genetic algorithm augmented with a support vector machine to automatically find a superior deep learning network topology and hyper-parameter set than a human expert can find in months. For the application of electron microscopy, the system furthers the goal of improving our understanding of the electron-beam-matter interactions and real-time image-based feedback, which enables a huge step beyond human capacity towards nanofabricating materials automatically. MENNDL has been scaled to the 4,200 available nodes of Summit achieving a measured 152.5 PFlops, with an estimated sustained performance of 167 PFlops when the entire machine is available.

Patton, Robert↗
NASA NTRS2008

Pre-Launch Algorithms and Risk Reduction in Support of the Geostationary Lightning Mapper for GOES-R and Beyond

The Geostationary Lightning Mapper (GLM) is a single channel, near-IR optical transient event detector, used to detect, locate and measure total lightning activity over the full-disk as part of a 3-axis stabilized, geostationary weather satellite system. The next generation NOAA Geostationary Operational Environmental Satellite (GOES-R) series with a planned launch in 2014 will carry a GLM that will provide continuous day and night observations of lightning from the west coast of Africa (GOES-E) to New Zealand (GOES-W) when the constellation is fully operational.The mission objectives for the GLM are to 1) provide continuous,full-disk lightning measurements for storm warning and Nowcasting, 2) provide early warning of tornadic activity, and 3) accumulate a long-term database to track decadal changes of lightning. The GLM owes its heritage to the NASA Lightning Imaging Sensor (1997-Present) and the Optical Transient Detector (1995-2000), which were developed for the Earth Observing System and have produced a combined 13 year data record of global lightning activity. Instrument formulation studies were completed in March 2007 and the implementation phase to develop a prototype model and up to four flight units is expected to begin in latter part of the year. In parallel with the instrument development, a GOES-R Risk Reduction Team and Algorithm Working Group Lightning Applications Team have begun to develop the Level 2B algorithms and applications. Proxy total lightning data from the NASA Lightning Imaging Sensor on the Tropical Rainfall Measuring Mission (TRMM) sate]lite and regional test beds (e.g., Lightning Mapping Arrays in North Alabama and the Washington DC Metropolitan area) are being used to develop the pre-launch algorithms and applications, and also improve our knowledge of thunderstorm initiation and evolution. Real time lightning mapping data provided to selected National Weather Service forecast offices in Southern and Eastern Region are also improving our understanding of the application of these data in the severe storm warning process and help to accelerate the development of the pre-launch algorithms and Nowcasting applications. Abstract for the 3 rd Conference on Meteorological

Goodman, Steven↗
NASA NTRS2008

Pre-Launch Algorithms and Risk Reduction in Support of the Geostationary Lightning Mapper for GOES-R and Beyond

The Geostationary Lightning Mapper (GLM) is a single channel, near-IR optical transient event detector, used to detect, locate and measure total lightning activity over the full-disk as part of a 3-axis stabilized, geostationary weather satellite system. The next generation NOAA Geostationary Operational Environmental Satellite (GOES-R) series with a planned launch in 2014 will carry a GLM that will provide continuous day and night observations of lightning from the west coast of Africa (GOES-E) to New Zealand (GOES-W) when the constellation is fully operational. The mission objectives for the GLM are to 1) provide continuous,full-disk lightning measurements for storm warning and Nowcasting, 2) provide early warning of tornado activity, and 3) accumulate a long-term database to track decadal changes of lightning. The GLM owes its heritage to the NASA Lightning Imaging Sensor (1997-Present) and the Optical Transient Detector (1995-2000), which were developed for the Earth Observing System and have produced a combined 13 year data record of global lightning activity. Instrument formulation studies were completed in March 2007 and the implementation phase to develop a prototype model and up to four flight units is expected to begin in latter part of the year. In parallel with the instrument development, a GOES-R Risk Reduction Team and Algorithm Working Group Lightning Applications Team have begun to develop the Level 2B algorithms and applications. Proxy total lightning data from the NASA Lightning Imaging Sensor on the Tropical Rainfall Measuring Mission (TRMM) satellite and regional test beds (e.g., Lightning Mapping Arrays in North Alabama and the Washington DC Metropolitan area) are being used to develop the pre-launch algorithms and applications, and also improve our knowledge of thunderstorm initiation and evolution. Real time lightning mapping data provided to selected National Weather Service forecast offices in Southern and Eastern Region are also improving our understanding of the application of these data in the severe storm warning process and help to accelerate the development of the pre-launch algorithms and Nowcasting applications.

Goodman, Steven↗
DOE OSTI2022

Profiling the BLAST bioinformatics application for load balancing on high-performance computing clusters

Abstract Background The Basic Local Alignment Search Tool (BLAST) is a suite of commonly used algorithms for identifying matches between biological sequences. The user supplies a database file and query file of sequences for BLAST to find identical sequences between the two. The typical millions of database and query sequences make BLAST computationally challenging but also well suited for parallelization on high-performance computing clusters. The efficacy of parallelization depends on the data partitioning, where the optimal data partitioning relies on an accurate performance model. In previous studies, a BLAST job was sped up by 27 times by partitioning the database and query among thousands of processor nodes. However, the optimality of the partitioning method was not studied. Unlike BLAST performance models proposed in the literature that usually have problem size and hardware configuration as the only variables, the execution time of a BLAST job is a function of database size, query size, and hardware capability. In this work, the nucleotide BLAST application BLASTN was profiled using three methods: shell-level profiling with the Unix “time” command, code-level profiling with the built-in “profiler” module, and system-level profiling with the Unix “gprof” program. The runtimes were measured for six node types, using six different database files and 15 query files, on a heterogeneous HPC cluster with 500+ nodes. The empirical measurement data were fitted with quadratic functions to develop performance models that were used to guide the data parallelization for BLASTN jobs. Results Profiling results showed that BLASTN contains more than 34,500 different functions, but a single function, RunMTBySplitDB, takes 99.12% of the total runtime. Among its 53 child functions, five core functions were identified to make up 92.12% of the overall BLASTN runtime. Based on the performance models, static load balancing algorithms can be applied to the BLASTN input data to minimize the runtime of the longest job on an HPC cluster. Four test cases being run on homogeneous and heterogeneous clusters were tested. Experiment results showed that the runtime can be reduced by 81% on a homogeneous cluster and by 20% on a heterogeneous cluster by re-distributing the workload. Discussion Optimal data partitioning can improve BLASTN’s overall runtime 5.4-fold in comparison with dividing the database and query into the same number of fragments. The proposed methodology can be used in the other applications in the BLAST+ suite or any other application as long as source code is available.

59 BASIC BIOLOGICAL SCIENCES↗