Search NASA⌕ Search

SEARCH · Search NASA

Results for “data dependencies”

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 73 records · Page 4

Benchmarking Variables for Checkpointing in HPC Applications

Checkpoint/Restart (C/R) is a widely used fault tolerance mechanism in converged systems of cloud, edge, and HPC. However, users often rely on their experience to determine which variables to checkpoint, as there is currently no benchmark that can provide a reference. This can result in checkpointing redundant or even incorrect variables. To address this issue, we propose a benchmark suite that includes critical variables for checkpointing, which have been manually identified, and a method for identifying those critical variables, with 20 representative HPC applications. Our method involves analyzing data dependency between variables to identify critical variables analytically. We verify the identified variables' correctness with a widely used C/R library FTI by an ablation study. With our benchmark suite and data dependency analysis, HPC practitioners now have a reference for identifying checkpointing variables and better knowledge of what kind of variables to checkpoint.

Fu, Xiang↗

(U) Physics Validation of the RMI-Based Ejecta Source Model Implementation in FLAG: L2 Milestone #6035 Report

The Los Alamos Physics and Engineering Models (PEM) program has developed a model for Richtmyer-Meshkov instability (RMI) based ejecta production from shock-melted surfaces, along with a prescription for a self-similar velocity distribution (SSVD) of the resulting ejecta particles. We have undertaken an effort to validate this source model using data from explosively driven tin coupon experiments. The model’s current formulation lacks a crucial piece of physics: a method for determining the duration of the ejecta production interval. Without a mechanism for terminating ejecta production, the model is not predictive. Furthermore, when the production interval is hand-tuned to match time-integrated mass data, the predicted time-dependent mass accumulation on a downstream sensor rises too sharply at early times and too slowly at late times because the SSVD overestimates the amount of mass stored in the fastest particles and underestimates the mass stored in the slowest particles. The functional form of the resulting m(t) is inconsistent with the available time-dependent data; numerical simulations and analytic studies agree on this point. Simulated mass tallies are highly sensitive to radial expansion of the ejecta cloud. It is not clear if the same effect is present in the experimental data but if so, depending on the degree, this may challenge the model’s compatibility with tin coupon data. The current implementation of the model in FLAG is sensitive to the detailed interaction between kinematics (hydrodynamic methods) and thermodynamics (material models); this sensitivity prohibits certain physics modeling choices. The appendices contain an extensive analytic study of piezoelectric ejecta mass measurements, along with test problems, excerpted from a longer work (LA-UR-17-21218).

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Computer-Aided Parallelizer and Optimizer

The Computer-Aided Parallelizer and Optimizer (CAPO) automates the insertion of compiler directives (see figure) to facilitate parallel processing on Shared Memory Parallel (SMP) machines. While CAPO currently is integrated seamlessly into CAPTools (developed at the University of Greenwich, now marketed as ParaWise), CAPO was independently developed at Ames Research Center as one of the components for the Legacy Code Modernization (LCM) project. The current version takes serial FORTRAN programs, performs interprocedural data dependence analysis, and generates OpenMP directives. Due to the widely supported OpenMP standard, the generated OpenMP codes have the potential to run on a wide range of SMP machines. CAPO relies on accurate interprocedural data dependence information currently provided by CAPTools. Compiler directives are generated through identification of parallel loops in the outermost level, construction of parallel regions around parallel loops and optimization of parallel regions, and insertion of directives with automatic identification of private, reduction, induction, and shared variables. Attempts also have been made to identify potential pipeline parallelism (implemented with point-to-point synchronization). Although directives are generated automatically, user interaction with the tool is still important for producing good parallel codes. A comprehensive graphical user interface is included for users to interact with the parallelization process.

Jin, Haoqiang↗

Sequential Principal Component Analysis -An Optimal and Hardware-Implementable Transform for Image Compression

This paper presents the JPL-developed Sequential Principal Component Analysis (SPCA) algorithm for feature extraction / image compression, based on "dominant-term selection" unsupervised learning technique that requires an order-of-magnitude lesser computation and has simpler architecture compared to the state of the art gradient-descent techniques. This algorithm is inherently amenable to a compact, low power and high speed VLSI hardware embodiment. The paper compares the lossless image compression performance of the JPL's SPCA algorithm with the state of the art JPEG2000, widely used due to its simplified hardware implementability. JPEG2000 is not an optimal data compression technique because of its fixed transform characteristics, regardless of its data structure. On the other hand, conventional Principal Component Analysis based transform (PCA-transform) is a data-dependent-structure transform. However, it is not easy to implement the PCA in compact VLSI hardware, due to its highly computational and architectural complexity. In contrast, the JPL's "dominant-term selection" SPCA algorithm allows, for the first time, a compact, low-power hardware implementation of the powerful PCA algorithm. This paper presents a direct comparison of the JPL's SPCA versus JPEG2000, incorporating the Huffman and arithmetic coding for completeness of the data compression operation. The simulation results show that JPL's SPCA algorithm is superior as an optimal data-dependent-transform over the state of the art JPEG2000. When implemented in hardware, this technique is projected to be ideally suited to future NASA missions for autonomous on-board image data processing to improve the bandwidth of communication.

Duong, Tuan A.↗

Clustering execution in a processing system to increase power savings

Embodiments relate to clustering execution in a processing system. An aspect includes accessing a control flow graph that defines a data dependency and an execution sequence of a plurality of tasks of an application that executes on a plurality of system components. The execution sequence of the tasks in the control flow graph is modified as a clustered control flow graph that clusters active and idle phases of a system component while maintaining the data dependency. The clustered control flow graph is sent to an operating system, where the operating system utilizes the clustered control flow graph for scheduling the tasks.

97 MATHEMATICS AND COMPUTING↗

Clustering execution in a processing system to increase power savings

Embodiments relate to clustering execution in a processing system. An aspect includes accessing a control flow graph that defines a data dependency and an execution sequence of a plurality of tasks of an application that executes on a plurality of system components. The execution sequence of the tasks in the control flow graph is modified as a clustered control flow graph that clusters active and idle phases of a system component while maintaining the data dependency. The clustered control flow graph is sent to an operating system, where the operating system utilizes the clustered control flow graph for scheduling the tasks.

97 MATHEMATICS AND COMPUTING↗

Use Computer-Aided Tools to Parallelize Large CFD Applications

Porting applications to high performance parallel computers is always a challenging task. It is time consuming and costly. With rapid progressing in hardware architectures and increasing complexity of real applications in recent years, the problem becomes even more sever. Today, scalability and high performance are mostly involving handwritten parallel programs using message-passing libraries (e.g. MPI). However, this process is very difficult and often error-prone. The recent reemergence of shared memory parallel (SMP) architectures, such as the cache coherent Non-Uniform Memory Access (ccNUMA) architecture used in the SGI Origin 2000, show good prospects for scaling beyond hundreds of processors. Programming on an SMP is simplified by working in a globally accessible address space. The user can supply compiler directives, such as OpenMP, to parallelize the code. As an industry standard for portable implementation of parallel programs for SMPs, OpenMP is a set of compiler directives and callable runtime library routines that extend Fortran, C and C++ to express shared memory parallelism. It promises an incremental path for parallel conversion of existing software, as well as scalability and performance for a complete rewrite or an entirely new development. Perhaps the main disadvantage of programming with directives is that inserted directives may not necessarily enhance performance. In the worst cases, it can create erroneous results. While vendors have provided tools to perform error-checking and profiling, automation in directive insertion is very limited and often failed on large programs, primarily due to the lack of a thorough enough data dependence analysis. To overcome the deficiency, we have developed a toolkit, CAPO, to automatically insert OpenMP directives in Fortran programs and apply certain degrees of optimization. CAPO is aimed at taking advantage of detailed inter-procedural dependence analysis provided by CAPTools, developed by the University of Greenwich, to reduce potential errors made by users. Earlier tests on NAS Benchmarks and ARC3D have demonstrated good success of this tool. In this study, we have applied CAPO to parallelize three large applications in the area of computational fluid dynamics (CFD): OVERFLOW, TLNS3D and INS3D. These codes are widely used for solving Navier-Stokes equations with complicated boundary conditions and turbulence model in multiple zones. Each one comprises of from 50K to 1,00k lines of FORTRAN77. As an example, CAPO took 77 hours to complete the data dependence analysis of OVERFLOW on a workstation (SGI, 175MHz, R10K processor). A fair amount of effort was spent on correcting false dependencies due to lack of necessary knowledge during the analysis. Even so, CAPO provides an easy way for user to interact with the parallelization process. The OpenMP version was generated within a day after the analysis was completed. Due to sequential algorithms involved, code sections in TLNS3D and INS3D need to be restructured by hand to produce more efficient parallel codes. An included figure shows preliminary test results of the generated OVERFLOW with several test cases in single zone. The MPI data points for the small test case were taken from a handcoded MPI version. As we can see, CAPO's version has achieved 18 fold speed up on 32 nodes of the SGI O2K. For the small test case, it outperformed the MPI version. These results are very encouraging, but further work is needed. For example, although CAPO attempts to place directives on the outer- most parallel loops in an interprocedural framework, it does not insert directives based on the best manual strategy. In particular, it lacks the support of parallelization at the multi-zone level. Future work will emphasize on the development of methodology to work in a multi-zone level and with a hybrid approach. Development of tools to perform more complicated code transformation is also needed.

Jin, H.↗

Temperature-dependent optical measurements of the dominant recombination mechanisms in InAs/InAsSb type-2 superlattices

Temperature-dependent measurements of carrier recombination rates using a time-resolved optical pump-probe technique are reported for mid-wave infrared InAs/InAs{sub 1−x}Sb{sub x} type-2 superlattices (T2SLs). By engineering the layer widths and alloy compositions, a 16 K band-gap of ∼235 ± 10 meV was achieved for five unintentionally and four intentionally doped T2SLs. Carrier lifetimes were determined by fitting lifetime models based on Shockley-Read-Hall (SRH), radiative, and Auger recombination processes to the temperature and excess carrier density dependent data. The minority carrier (MC), radiative, and Auger lifetimes were observed to generally increase with increasing antimony content and decreasing layer thickness for the unintentionally doped T2SLs. The MC lifetime is limited by SRH processes at temperatures below 200 K in the unintentionally doped T2SLs. The extracted SRH defect energy levels were found to be near mid-bandgap. Also, it is observed that the MC lifetime is limited by Auger recombination in the intentionally doped T2SLs with doping levels greater than n ∼ 10{sup 16} cm{sup −3}.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Efficient Machine Learning Approach for Optimizing Scientific Computing Applications on Emerging HPC Architectures

Efficient parallel implementations of scientific applications on multi-core CPUs with accelerators such as GPUs and Xeon Phis is challenging. This requires - exploiting the data parallel architecture of the accelerator along with the vector pipelines of modern x86 CPU architectures, load balancing, and efficient memory transfer between different devices. It is relatively easy to meet these requirements for highly structured scientific applications. In contrast, a number of scientific and engineering applications are unstructured. Getting performance on accelerators for these applications is extremely challenging because many of these applications employ irregular algorithms which exhibit data-dependent control-ow and irregular memory accesses. Furthermore, these applications are often iterative with dependency between steps, and thus making it hard to parallelize across steps. As a result, parallelism in these applications is often limited to a single step. Numerical simulation of charged particles beam dynamics is one such application where the distribution of work and memory access pattern at each time step is irregular. Applications with these properties tend to present significant branch and memory divergence, load imbalance between different processor cores, and poor compute and memory utilization. Prior research on parallelizing such irregular applications have been focused around optimizing the irregular, data-dependent memory accesses and control-ow during a single step of the application independent of the other steps, with the assumption that these patterns are completely unpredictable. We observed that the structure of computation leading to control-ow divergence and irregular memory accesses in one step is similar to that in the next step. It is possible to predict this structure in the current step by observing the computation structure of previous steps. In this dissertation, we present novel machine learning based optimization techniques to address the parallel implementation challenges of such irregular applications on different HPC architectures. In particular, we use supervised learning to predict the computation structure and use it to address the control-ow and memory access irregularities in the parallel implementation of such applications on GPUs, Xeon Phis, and heterogeneous architectures composed of multi-core CPUs with GPUs or Xeon Phis. We use numerical simulation of charged particles beam dynamics simulation as a motivating example throughout the dissertation to present our new approach, though they should be equally applicable to a wide range of irregular applications. The machine learning approach presented here use predictive analytics and forecasting techniques to adaptively model and track the irregular memory access pattern at each time step of the simulation to anticipate the future memory access pattern. Access pattern forecasts can then be used to formulate optimization decisions during application execution which improves the performance of the application at a future time step based on the observations from earlier time steps. In heterogeneous architectures, forecasts can also be used to improve the memory performance and resource utilization of all the processing units to deliver a good aggregate performance. We used these optimization techniques and anticipation strategy to design a cache-aware, memory efficient parallel algorithm to address the irregularities in the parallel implementation of charged particles beam dynamics simulation on different HPC architectures. Experimental result using a diverse mix of HPC architectures shows that our approach in using anticipation strategy is effective in maximizing data reuse, ensuring workload balance, minimizing branch and memory divergence, and in improving resource utilization.

Arumugam, Kamesh↗

Lunar Relay Satellite Network for Space Exploration: Architecture, Technologies and Challenges

NASA is planning a series of short and long duration human and robotic missions to explore the Moon and then Mars. A key objective of these missions is to grow, through a series of launches, a system of systems infrastructure with the capability for safe and sustainable autonomous operations at minimum cost while maximizing the exploration capabilities and science return. An incremental implementation process will enable a buildup of the communication, navigation, networking, computing, and informatics architectures to support human exploration missions in the vicinities and on the surfaces of the Moon and Mars. These architectures will support all space and surface nodes, including other orbiters, lander vehicles, humans in spacesuits, robots, rovers, human habitats, and pressurized vehicles. This paper describes the integration of an innovative MAC and networking technology with an equally innovative position-dependent, data routing, network technology. The MAC technology provides the relay spacecraft with the capability to autonomously discover neighbor spacecraft and surface nodes, establish variable-rate links and communicate simultaneously with multiple in-space and surface clients at varying and rapidly changing distances while making optimum use of the available power. The networking technology uses attitude sensors, a time synchronization protocol and occasional orbit-corrections to maintain awareness of its instantaneous position and attitude in space as well as the orbital or surface location of its communication clients. A position-dependent data routing capability is used in the communication relay satellites to handle the movement of data among any of multiple clients (including Earth) that may be simultaneously in view; and if not in view, the relay will temporarily store the data from a client source and download it when the destination client comes into view. The integration of the MAC and data routing networking technologies would enable a relay satellite system to provide end-to-end communication services for robotic and human missions in the vicinity, or on the surface of the Moon with a minimum of Earth-based operational support.

Bhasin, Kul B.↗

Parallelization of implicit finite difference schemes in computational fluid dynamics

Implicit finite difference schemes are often the preferred numerical schemes in computational fluid dynamics, requiring less stringent stability bounds than the explicit schemes. Each iteration in an implicit scheme involves global data dependencies in the form of second and higher order recurrences. Efficient parallel implementations of such iterative methods are considerably more difficult and non-intuitive. The parallelization of the implicit schemes that are used for solving the Euler and the thin layer Navier-Stokes equations and that require inversions of large linear systems in the form of block tri-diagonal and/or block penta-diagonal matrices is discussed. Three-dimensional cases are emphasized and schemes that minimize the total execution time are presented. Partitioning and scheduling schemes for alleviating the effects of the global data dependencies are described. An analysis of the communication and the computation aspects of these methods is presented. The effect of the boundary conditions on the parallel schemes is also discussed.

Decker, Naomi H.↗

Symmetric Indefinite Linear Solver Using OpenMP Task on Multicore Architectures

Recently, the Open Multi-Processing (OpenMP) standard has incorporated task-based programming, where a function call with input and output data is treated as a task. At run time, OpenMP's superscalar scheduler tracks the data dependencies among the tasks and executes the tasks as their dependencies are resolved. On a shared-memory architecture with multiple cores, the independent tasks are executed on different cores in parallel, thereby enabling parallel execution of a seemingly sequential code. With the emergence of many-core architectures, this type of programming paradigm is gaining attention-not only because of its simplicity, but also because it breaks the artificial synchronization points of the program and improves its thread-level parallelization. In this paper, we use these new OpenMP features to develop a portable high-performance implementation of a dense symmetric indefinite linear solver. Obtaining high performance from this kind of solver is a challenge because the symmetric pivoting, which is required to maintain numerical stability, leads to data dependencies that prevent us from using some common performance-improving techniques. To fully utilize a large number of cores through tasking, while conforming to the OpenMP standard, we describe several techniques. Our performance results on current many-core architectures-including Intel's Broadwell, Intel's Knights Landing, IBM's Power8, and Arm's ARMv8-demonstrate the portable and superior performance of our implementation compared with the Linear Algebra PACKage (LAPACK). The resulting solver is now available as a part of the PLASMA software package.

Computer Science↗

Multi-fluid transport code modeling of time-dependent recycling in ELMy H-mode

Simulations of a high-confinement-mode (H-mode) tokamak discharge with infrequent giant type-I ELMs are performed by the multi-fluid, multi-species, two-dimensional transport code UEDGE-MB, which incorporates the Macro-Blob approach for intermittent non-diffusive transport due to filamentary coherent structures observed during the Edge Localized Modes (ELMs) and simple time-dependent multi-parametric models for cross-field plasma transport coefficients and working gas inventory in material surfaces. Temporal evolutions of pedestal plasma profiles, divertor recycling, and wall inventory in a sequence of ELMs are studied and compared to the experimental time-dependent data. Short- and long-time-scale variations of the pedestal and divertor plasmas where the ELM is described as a sequence of macro-blobs are discussed. It is shown that the ELM recovery includes the phase of relatively dense and cold post-ELM divertor plasma evolving on a several ms scale, which is set by the transport properties of H-mode barrier. The global gas balance in the discharge is also analyzed. The calculated rates of working gas deposition during each ELM and wall outgassing between ELMs are compared to the ELM particle losses from the pedestal and neutral-beam-injection fueling rate, correspondingly. A sensitivity study of the pedestal and divertor plasmas to model assumptions for gas deposition and release on material surfaces is presented. The performed simulations show that the dynamics of pedestal particle inventory is dominated by the transient intense gas deposition into the wall during each ELM followed by continuous gas release between ELMs at roughly a constant rate.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A modified moment-fitted integration scheme for X-FEM applications with history-dependent material data

Here, we present a strategy for the numerical integration of partial elements with the eXtended finite element method (X-FEM). The new strategy is specifically designed for problems with propagating cracks through a bulk material that exhibits inelasticity. Following a standard approach with the X-FEM, as the crack propagates into new regions of the domain, elements are split into several copies that contain pieces of the domain on either side of the crack. We examine quadrature rules that have sufficient accuracy to calculate stiffness matrices regardless of the orientation of the crack with respect to the element. This permits the number of integration points within elements to remain constant as a crack propagates, and for state data to be easily transferred between successive discretizations. In order to maintain weights that are strictly positive, we propose an approach that blends moment-fitted weights with volume-fraction based weights. To demonstrate the efficacy of this simple approach, we present results from numerical tests and examples with both elastic and plastic material response.

42 ENGINEERING↗

Challenges of Monte Carlo Transport

These are slides from a presentation for Parallel Summer School at Los Alamos National Laboratory. Solving discretized partial differential equations (PDEs) of interest can require a large number of computations. We can identify concurrency to allow parallel solution of discrete PDEs. Simulated particles histories can be used to solve the Boltzmann transport equation. Particle histories are independent in neutral particle transport, making them amenable to parallel computation. Physical parameters and method type determine the data dependencies of particle histories. Data requirements shape parallel algorithms for Monte Carlo. Then, Parallel Computational Physics and Parallel Monte Carlo are discussed and, finally, the results are given. The mesh passing method greatly simplifies the IMC implementation and allows simple load-balancing. Using MPI windows and passive, one-sided RMA further simplifies the implementation by removing target synchronization. The author is very interested in implementations of PGAS that may allow further optimization for one-sided, read-only memory access (e.g. Open SHMEM). The MPICH_RMA_OVER_DMAPP option and library is required to make one-sided messaging scale on Trinitite - Moonlight scales poorly. Interconnect specific libraries or functions are likely necessary to ensure performance. BRANSON has been used to directly compare the current standard method to a proposed method on idealized problems. The mesh passing algorithm performs well on problems that are designed to show the scalability of the particle passing method. BRANSON can now run load-imbalanced, dynamic problems. Potential avenues of improvement in the mesh passing algorithm will be implemented and explored. A suite of test problems that stress DD methods will elucidate a possible path forward for production codes.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Data Analysis Approaches for the Risk-Informed Safety Margins Characterization Toolkit

In the past decades, several numerical simulation codes have been employed to simulate accident dynamics (e.g., RELAP5-3D, RELAP-7, MELCOR, MAAP). In order to evaluate the impact of uncertainties into accident dynamics, several stochastic methodologies have been coupled with these codes. These stochastic methods range from classical Monte-Carlo and Latin Hypercube sampling to stochastic polynomial methods. Similar approaches have been introduced into the risk and safety community where stochastic methods (such as RAVEN, ADAPT, MCDET, ADS) have been coupled with safety analysis codes in order to evaluate the safety impact of timing and sequencing of events. These approaches are usually called Dynamic PRA or simulation-based PRA methods. These uncertainties and safety methods usually generate a large number of simulation runs (database storage may be on the order of gigabytes or higher). The scope of this paper is to present a broad overview of methods and algorithms that can be used to analyze and extract information from large data sets containing time dependent data. In this context, “extracting information” means constructing input-output correlations, finding commonalities, and identifying outliers. Some of the algorithms presented here have been developed or are under development within the RAVEN statistical framework.

97 MATHEMATICS AND COMPUTING↗

Compiler and Runtime Approaches to Enable Large-Scale Irregular Programs. Final report, July 2013 - July 2019

While regular algorithms, characterized by operations on dense matrices and arrays, have long been the mainstay of scientific, high-performance computing, irregular algorithms, which feature unpredictable accesses to pointer-based data structures, are becoming increasingly common in high performance computing, arising in graph analysis, data mining and visualization, among other domains. Unfortunately, the defining characteristics of irregular applications, their dynamic, unpredictable, data-dependent access patterns and data layouts, make achieving high performance on large scale systems difficult. Scaling applications to peta- and exa-scale requires carefully controlling communication and data movement and placement, an inherently difficult task when access patterns and data layouts are unpredictable! Most irregular applications that attain high performance must be painstakingly hand-written and hand-tuned, with few common principles or paradigms uniting various implementations and easing future development. Despite the increasing importance of irregular applications, there is little programmer knowledge, and even less compiler ability, devoted to optimizing them. This project aims to solve these problems. By allowing programmers to write irregular applications in high level forms, with at most a few annotations highlighting key structural properties, programmers can focus on developing their algorithms and methods. The compiler and run-time system can take on the tedious task of optimizing the application for execution at large scales, and can automatically provide efficient implementations. This will provide portability and ease maintenance for existing irregular applications, but, more importantly, open up whole new domains of computational science to large-scale, high-performance simulation codes.

97 MATHEMATICS AND COMPUTING↗