Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel computers.”

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 343 records · Page 19

MapReduce MPI library (MR-MPI) v. 1.0

The MapReduce MPI library is a software tool for performing MapReduce operations on a distributed memory parallel computer via message passing (MPI). These are data-processing or computational operations which achieve parallelism by breaking a large task into two stages, a map and a reduce . Each of these are formulated as simple on-processor functions which the user can easily write. The library assigns independent tasks to processors and performs the parallel data movement and transformations behind the scenes.

PLIMPTON, STEVEN↗

A GPU‐Accelerated Generative Adversarial Model for Causal Inference

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

GPU↗

Early Evaluation of Directive-Based GPU Programming Models for Productive Exascale Computing

Graphics Processing Unit (GPU)-based parallel computer architectures have shown increased popularity as a building block for high performance computing, and possibly for future Exascale computing. However, their programming complexity remains as a major hurdle for their widespread adoption. To provide better abstractions for programming GPU architectures, researchers and vendors have proposed several directive-based GPU programming models. These directive-based models provide different levels of abstraction, and required different levels of programming effort to port and optimize applications. Understanding these differences among these new models provides valuable insights on their applicability and performance potential. In this paper, we evaluate existing directive-based models by porting thirteen application kernels from various scientific domains to use CUDA GPUs, which, in turn, allows us to identify important issues in the functionality, scalability, tunability, and debuggability of the existing models. Our evaluation shows that directive-based models can achieve reasonable performance, compared to hand-written GPU codes.

Lee, Seyong↗

Skel: Generative Software for Producing Skeletal I/O Applications

Massively parallel computations consist of a mixture of computation, communication, and I/O. As part of the co-design for the inevitable progress towards exascale computing, we must apply lessons learned from past work to succeed in this new age of computing. Of the three components listed above, implementing an effective parallel I/O solution has often been overlooked by application scientists and was usually added to large scale simulations only when existing serial techniques had failed. As scientists teams scaled their codes to run on hundreds of processors, it was common to call on an I/O expert to implement a set of more scalable I/O routines. These routines were easily separated from the calculations and communication, and in many cases, an I/O kernel was derived from the application which could be used for testing I/O performance independent of the application. These I/O kernels developed a life of their own used as a broad measure for comparing different I/O techniques. Unfortunately, as years passed and computation and communication changes required changes to the I/O, the separate I/O kernel used for benchmarking remained static no longer providing an accurate indicator of the I/O performance of the simulation making I/O research less relevant for the application scientists. In this paper we describe a new approach to this problem where I/O kernels are replaced with skeletal I/O applications automatically generated from an abstract set of simulation I/O parameters. We realize this abstraction by leveraging the ADIOS middleware's XML I/O specification with additional runtime parameters. Skeletal applications offer all of the benefits of I/O kernels including allowing I/O optimizations to focus on useful I/O patterns. Moreover, since they are automatically generated, it is easy to produce an updated I/O skeleton whenever the simulation's I/O changes. In this paper we analyze the performance of automatically generated I/O skeletal applications for the S3D and GTS codes. We show that these skeletal applications achieve performance comparable to that of the production applications. We wrap up the paper with a discussion of future changes to make the skeletal application better approximate the actual I/O performed in the simulation.

99 GENERAL AND MISCELLANEOUS↗

GPU-acceleration of the ELPA2 distributed eigensolver for dense symmetric and hermitian eigenproblems

The solution of eigenproblems is often a key computational bottleneck that limits the tractable system size of numerical algorithms, among them electronic structure theory in chemistry and in condensed matter physics. Large eigenproblems can easily exceed the capacity of a single compute node, thus must be solved on distributed-memory parallel computers. We here present GPU-oriented optimizations of the ELPA two-stage tridiagonalization eigensolver (ELPA2). On top of cuBLAS-based GPU offloading, we add a CUDA kernel to speed up the back-transformation of eigenvectors, which can be the computationally most expensive part of the two-stage tridiagonalization algorithm. Furthermore, we benchmark the performance of this GPU-accelerated eigensolver on two hybrid CPU–GPU architectures, namely a compute cluster based on Intel Xeon Gold CPUs and NVIDIA Volta GPUs, and the Summit supercomputer based on IBM POWER9 CPUs and NVIDIA Volta GPUs. Consistent with previous benchmarks on CPU-only architectures, the GPU-accelerated two-stage solver exhibits a parallel performance superior to the one-stage counterpart. Finally, we demonstrate the performance of the GPU-accelerated eigensolver developed in this work for routine semi-local KS-DFT calculations comprising thousands of atoms.

97 MATHEMATICS AND COMPUTING↗

Data Processing for the SNS EQ-SANS Diffractometer

The scattering data on the Extended Q-Range Small Angle Scattering Diffractometer at the Spallation Neutron Source are stored as neutron events. Each event has two parts: the time of fight of the detected neutron, and an identifier for the neutron's location on the detector. In addition, each event is associated with the source pulse from which the neutron is originated. Due to the high neutron beam intensity at the SNS, the size of the stored data files can easily exceed several Gigabytes. Processing and reducing these data can be very time consuming. The standard way of handling the data at the SNS has been to first transform the data into histograms. On the EQ-SANS, these histograms are further processed into intensity I vs. neutron momentum transfer Q files, I(Q). For a large data, the whole process can take many minutes to hours to complete, which severely limits the turnaround time of an experiment. Experimenters often want to reduce the data quickly and depending on the result, make decisions on the setup of next measurements. They may also need to reduce the data using different inputs and conditions. In this work, we present the implementation of a fast data processing scheme for the EQ-SANS diffractometer. The scattering data are processed directly from neutron event files into I(Q) or other desired formats, greatly reducing the needed computation time. Further speed gains are obtained by the implementation of parallel computing under the Message Passing Interface framework. Most data on the EQ-SANS can now be handled in mere seconds. The process is highly automated. All relevant experimental information that is stored in instrument configuration files are read and processed. In addition, the new implementation is highly flexible. All the data corrections and handlings can be controlled transparently through user inputs. Finally, software implementation enables the slicing of neutron event, which can be used for the study of time dependent phenomena.

Data Processing↗

Quantum Package 2.0: An Open-Source Determinant-Driven Suite of Programs

Quantum chemistry is a discipline which relies heavily on very expensive numerical computations. The scaling of correlated wave function methods lies, in their standard implementation, between O(N-5) and O(e(N)), where N is proportional to the system size. Therefore, performing accurate calculations on chemically meaningful systems requires (i) approximations that can lower the computational scaling and (ii) efficient implementations that take advantage of modern massively parallel architectures. QUANTUM PACKAGE is an open-source programming environment for quantum chemistry specially designed for wave function methods. Its main goal is the development of determinant-driven selected configuration interaction (sCI) methods and multireference second-order perturbation theory (PT2). The determinant-driven framework allows the programmer to include any arbitrary set of determinants in the reference space, hence providing greater methodological freedom. The sCI method implemented in QUANTUM PACKAGE is based on the CIPSI (Configuration Interaction using a Perturbative Selection made Iteratively) algorithm which complements the variational sCI energy with a PT2 correction. Additional external plugins have been recently added to perform calculations with multireference coupled cluster theory and range-separated density-functional theory. All the programs are developed with the IRPF90 code generator, which simplifies collaborative work and the development of new features. QUANTUM PACKAGE strives to allow easy implementation and experimentation of new methods, while making parallel computation as simple and efficient as possible on modern supercomputer architectures. Currently, the code enables, routinely, to realize runs on roughly 2 000 CPU cores, with tens of millions of determinants in the reference space. Moreover, we have been able to push up to 12 288 cores in order to test its parallel efficiency. In the present manuscript, we also introduce some key new developments: (i) a renormalized second-order perturbative correction for efficient extrapolation to the full CI limit and (ii) a stochastic version of the CIPSI selection performed simultaneously to the PT2 calculation at no extra cost.

Garniron, Yann↗

NWChem

Computational atomistic modeling has emerged as a powerful approach for fundamental studies in chemistry, materials science, geosciences, and biology. It has the ability to provide detailed insights into complex phenomena and over the years has grown into a powerful tool not only to validate experiment, but also to predict new phenomena. In order to harness computational modeling to the fullest, superior computational tools are required that not only offer a broad spectrum of capabilities, but are also capable of taking advantage of modern massively parallel computer architectures efficiently to solve real world problems. The NWChem program suite is an example along these lines. It offers a range of capabilities to perform first-principles electronic structure calculations on molecular and periodic systems, classical molecular dynamics simulations and hybrid QM/MM simulations that permits one to combine quantum and classical methods within a calculation. The code is designed to run on high-performance parallel supercomputers as well as conventional workstation clusters with the goal to provide scalable solutions for large scale atomistic simulations. It has been ported to almost all high-performance computing platforms, workstations, PCs running LINUX, as well as clusters of desktop platforms or workgroup servers. The package is scalable, both in its ability to treat large problems efficiently, and in its utilization of available parallel computing resources. The parallel framework for NWChem is provided by the Global Array (GA) toolkit developed at PNNL. NWChem is developed and maintained by the Environmental Molecular Sciences Laboratory (EMSL) located at the Pacific Northwest National Laboratory (PNNL) in Washington State. The latest release (version 6.0) of the code is distributed under the terms of the Educational Community License (ECL) version 2.0.

99 GENERAL AND MISCELLANEOUS↗

Machine and Application Aware Partitioning for Adaptive Mesh Refinement Applications. In: HPDC '17 Proceedings of the 26th International Symposium on High-Performance Parallel and Distributed Computing

Load balancing and partitioning are critical when it comes to parallel computations. Popular partitioning strategies based on space filling curves focus on equally dividing work. The partitions produced are independent of the architecture or the application. Given the ever-increasing relative cost of data movement and increasing heterogeneity of our architectures, it is no longer sufficient to only consider an equal partitioning of work. Minimizing communication costs are equally if not more important. Our hypothesis is that an unequal partitioning that minimizes communication costs significantly can scale and perform better than conventional equal-work partitioning schemes. This tradeoff is dependent on the architecture as well as the application. We validate our hypothesis in the context of a finite-element computation utilizing adaptive mesh-refinement. Our central contribution is a new partitioning scheme that minimizes the overall runtime of subsequent computations by performing architecture and application-aware non-uniform work assignment in order to decrease time to solution, primarily by minimizing data-movement. We evaluate our algorithm by comparing it against standard space-filling curve based partitioning algorithms and observing time-to-solution as well as energy-to-solution for solving Finite Element computations on adaptively refined meshes. We demonstrate excellent scalability of our new partition algorithm up to $262,144$ cores on ORNL's Titan and demonstrate that the proposed partitioning scheme reduces overall energy as well as time-to-solution for application codes by up to 22.0%.

Fernando, Milinda↗

International ACM Symposium on High Performance Parallel and Distributed Computing Conference for 2017, 2018, 2019, and 2020

The 28th ACM HPDC Conference was held in Phoenix, Arizona, June 24 and 28, 2019 (hpdc.org/2019), that was colocated with ACM FCRC 2019 (fcrc.acm.org). During the conference, Prof. Geoffrey Fox, Indiana University, was given the HPDC Achievement Award for 2019. Prof gave a keynote speech entitled “Perspectives on High-Performance Computing in a Big Data World. In addition, to the keynote speakers from HPDC and FCRC conferences, the conference organized successfully five workshops and one Ph.D. forum. The ACM FCRC had a total of 2700 attendees, and HPDC had a total of 120 attendees that included 32 students. We have used the DOE sponsorship to support the conference proceedings that acknowledge the DoE support and partially supported the travel to the HPDC PC meeting, Keynote speaker accommodation, best papers, presentation and poster award.

42 ENGINEERING↗

Improving MPI Collective I/O for High Volume Non-Contiguous Requests With Intra-Node Aggregation

Two-phase I/O is a well-known strategy for implementing collective MPI-IO functions. It redistributes I/O requests among the calling processes into a form that minimizes the file access costs. As modern parallel computers continue to grow into the exascale era, the communication cost of such request redistribution can quickly overwhelm collective I/O performance. In this study, this effect has been observed from parallel jobs that run on multiple compute nodes with a high count of MPI processes on each node. Furthermore, to reduce the communication cost, we present a new design for collective I/O by adding an extra communication layer that performs request aggregation among processes within the same compute nodes. This approach can significantly reduce inter-node communication contention when redistributing the I/O requests. We evaluate the performance and compare it with the original two-phase I/O on Cray XC40 parallel computers (Theta and Cori) with Intel KNL and Haswell processors. Using I/O patterns from two large-scale production applications and an I/O benchmark, we show our proposed method effectively reduces the communication cost and hence maintains the scalability for a large number of processes.

97 MATHEMATICS AND COMPUTING↗

Processing communications events in parallel active messaging interface by awakening thread from wait state

Processing data communications events in a parallel active messaging interface (`PAMI`) of a parallel computer that includes compute nodes that execute a parallel application, with the PAMI including data communications endpoints, and the endpoints are coupled for data communications through the PAMI and through other data communications resources, including determining by an advance function that there are no actionable data communications events pending for its context, placing by the advance function its thread of execution into a wait state, waiting for a subsequent data communications event for the context; responsive to occurrence of a subsequent data communications event for the context, awakening by the thread from the wait state; and processing by the advance function the subsequent data communications event now pending for the context.

97 MATHEMATICS AND COMPUTING↗

FLC Nomination: Stephen R. Lee

STEM Mentorship Award: Given to an individual who may or may not be a direct employee of a government laboratory or facility in the Mid Continent Region, but has demonstrated exceptional character, mentorship, and creativity in the areas of science and math or by promoting the development of young people in grades K-12 or higher in technical career fields. Contributions must be unique and show measurable benefit to the laboratory, school system/students and must have occurred within the past three calendar years. Stephen R. Lee is the division leader of the Computer, Computational, and Statistical Sciences Division (CCS) at Los Alamos National Laboratory (LANL or the Laboratory). He holds a Bachelor of Science in Physics from New Mexico State University and a Master of Science in Physics from Kansas State University and began his 27-year career at the Laboratory in 1986. As division leader of CCS, Stephen is responsible for the management and oversight of CCS and all of its activities, as well as approximately 200 technical staff engaged in computer and computational, and statistical sciences. His experience and interests include computational algorithm development for largescale parallel computing (peta-scale and beyond), predictive simulations for nuclear weapons, simulation capabilities for non-nuclear weapons applications such as bioscience, homeland defense, nanotechnology, advanced computational platform research and development.

97 MATHEMATICS AND COMPUTING↗

Dynamic model-based sensor network design algorithm for system efficiency maximization

Here, a dynamic model-based sensor network design (DMSND) algorithm has been developed for maximizing system efficiency for an estimator-based control system. The algorithm synthesizes the optimal sensor network in the face of disturbances or set point changes. Computational expense of the large-scale combinatorial optimization problem is significantly reduced by parallel computing and by using combination of three novel strategies: multi-rate sampling frequency, model order reduction, and use of an incumbent solution that enables early termination of evaluation of infeasible sensor sets. The developed algorithm is applied to an acid gas removal unit as part of an integrated gasification combined cycle power plant with carbon capture. Even though there are more than thousand process states and more than hundred candidate sensor locations, the optimal sensor network design problem for maximizing process efficiency could be solved within couple of hours for a given budget.

47 OTHER INSTRUMENTATION↗

Domain decomposition in time for PDE-constrained optimization

Here, PDE-constrained optimization problems have a wide range of applications, but they lead to very large and ill-conditioned linear systems, especially if the problems are time dependent. In this paper we outline an approach for dealing with such problems by decomposing them in time and applying an additive Schwarz preconditioner in time, so that we can take advantage of parallel computers to deal with the very large linear systems. We then illustrate the performance of our method on a variety of problems.

97 MATHEMATICS AND COMPUTING↗