Search NASA⌕ Search

SEARCH · Search NASA

Results for “asynchronous 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 145 records · Page 8

Collective network for computer structures

A system and method for enabling high-speed, low-latency global collective communications among interconnected processing nodes. The global collective network optimally enables collective reduction operations to be performed during parallel algorithm operations executing in a computer structure having a plurality of the interconnected processing nodes. Router devices are included that interconnect the nodes of the network via links to facilitate performance of low-latency global processing operations at nodes of the virtual network. The global collective network may be configured to provide global barrier and interrupt functionality in asynchronous or synchronized manner. When implemented in a massively-parallel supercomputing structure, the global collective network is physically and logically partitionable according to the needs of a processing algorithm.

97 MATHEMATICS AND COMPUTING↗

mystic - a brief introduction

The mystic framework enables the solving of difficult optimization problems through three major innovations: (1) plug-and-play component-based optimization algorithms, (2) the ability to dynamically construct and apply generalzed trnasforms, K , where x' = K(x) , and (3) workflows providing asynchronous execution and statefulness. These features, in conjunction with supporting packages pathos , pyina , and klepto , can provide massively-parallel scalable workflows for quickly solving optimization problems in complex nonlinear spaces. mystic can be used to simplify non-convex optimization problems by transforming away nonlinearities through user-built kernel transforms. This tutorial demonstrates using mystic for machine learning with automated dimensional-reduction, using embarrasingly parallel solver ensembles to find an accurate interpolated surrogate for a nonlinear surface, and the determination of worst-case bounds on expectation value of an objective function under uncertainty.

97 MATHEMATICS AND COMPUTING↗

A Parallel Saturation Algorithm on Shared Memory Architectures

Symbolic state-space generators are notoriously hard to parallelize. However, the Saturation algorithm implemented in the SMART verification tool differs from other sequential symbolic state-space generators in that it exploits the locality of ring events in asynchronous system models. This paper explores whether event locality can be utilized to efficiently parallelize Saturation on shared-memory architectures. Conceptually, we propose to parallelize the ring of events within a decision diagram node, which is technically realized via a thread pool. We discuss the challenges involved in our parallel design and conduct experimental studies on its prototypical implementation. On a dual-processor dual core PC, our studies show speed-ups for several example models, e.g., of up to 50% for a Kanban model, when compared to running our algorithm only on a single core.

Ezekiel, Jonathan↗

TBAA20: Task-Based Algorithms and Applications

The new challenges posed by Exascale system architectures have resulted in difficulty achieving a desired scalability using traditional distributed ­memory runtimes. Task­-based programming models show promise in addressing these challenges, providing application developers with a productive and performant approach to programming on next generation systems. Empirical studies show that task-based models can overcome load ­balancing issues that are inherent to traditional distributed ­memory runtimes, and that task-­based runtimes perform comparably to those systems when balanced. This panel is designed to explore the advantages of task-­based programming models on modern and future HPC systems from an industry, university, and national lab perspective. It aims at gathering application experts and proponents of these models to present concrete and practical examples of using task­-based runtimes to overcome the challenges posed by Exascale system architectures. This report describes the objectives, activities, and outcomes of the panel TBAA: Task­-Based Algorithms and Applications which was held at the International Conference for High Performance Computing, Networking, Storage, and Analysis (SC 20) on November 18, 2020.

97 MATHEMATICS AND COMPUTING↗

The OVIS analysis architecture.

This report summarizes the current statistical analysis capability of OVIS and how it works in conjunction with the OVIS data readers and interpolators. It also documents how to extend these capabilities. OVIS is a tool for parallel statistical analysis of sensor data to improve system reliability. Parallelism is achieved using a distributed data model: many sensors on similar components (metaphorically sheep) insert measurements into a series of databases on computers reserved for analyzing the measurements (metaphorically shepherds). Each shepherd node then processes the sheep data stored locally and the results are aggregated across all shepherds. OVIS uses the Visualization Tool Kit (VTK) statistics algorithm class hierarchy to perform analysis of each process's data but avoids VTK's model aggregation stage which uses the Message Passing Interface (MPI); this is because if a single process in an MPI job fails, the entire job will fail. Instead, OVIS uses asynchronous database replication to aggregate statistical models. OVIS has several additional features beyond those present in VTK that, first, accommodate its particular data format and, second, improve the memory and speed of the statistical analyses. First, because many statistical algorithms are multivariate in nature and sensor data is typically univariate, interpolation of data is required to provide simultaneous observations of metrics. Note that in this report, we will refer to a single value obtained from a sensor as a measurement while a collection of multiple sensor values simultaneously present in the system is an observation. A base class for interpolation is provided that abstracts the operation of converting multiple sensor measurements into simultaneous observations. A concrete implementation is provided that performs piecewise constant temporal interpolation of multiple metrics across a single component. Secondly, because calculations may summarize data too large to fit in memory OVIS analyses batches of observations at a time and aggregates these intermediate intra-process models as it goes before storing the final model for inter-process aggregation via database replication. This reduces the memory footprint of the analysis, interpolation, and the database client and server query processing. This also interleaves processing with the disk I/O required to fetch data from the database - also improving speed. This report documents how OVIS performs analyses and how to create additional analysis components that fetch measurements from the database, perform interpolation, or perform operations on streamed observations (such as model updates or assessments). The rest of this section outlines the OVIS analysis algorithm and is followed by sections specific to each subtask. Note that we are limiting our discussion for now to the creation of a model from a set of measurements, and not including the assessment of observations using a model. The same framework can be used for assessment but that use case is not detailed in this report.

97 MATHEMATICS AND COMPUTING↗

Exploring Gigabyte Datasets in Real Time: Architectures, Interfaces and Time-Critical Design

Architectures and Interfaces: The implications of real-time interaction on software architecture design: decoupling of interaction/graphics and computation into asynchronous processes. The performance requirements of graphics and computation for interaction. Time management in such an architecture. Examples of how visualization algorithms must be modified for high performance. Brief survey of interaction techniques and design, including direct manipulation and manipulation via widgets. talk discusses how human factors considerations drove the design and implementation of the virtual wind tunnel. Time-Critical Design: A survey of time-critical techniques for both computation and rendering. Emphasis on the assignment of a time budget to both the overall visualization environment and to each individual visualization technique in the environment. The estimation of the benefit and cost of an individual technique. Examples of the modification of visualization algorithms to allow time-critical control.

Bryson, Steve↗

Symbolic Computation of Strongly Connected Components Using Saturation

Finding strongly connected components (SCCs) in the state-space of discrete-state models is a critical task in formal verification of LTL and fair CTL properties, but the potentially huge number of reachable states and SCCs constitutes a formidable challenge. This paper is concerned with computing the sets of states in SCCs or terminal SCCs of asynchronous systems. Because of its advantages in many applications, we employ saturation on two previously proposed approaches: the Xie-Beerel algorithm and transitive closure. First, saturation speeds up state-space exploration when computing each SCC in the Xie-Beerel algorithm. Then, our main contribution is a novel algorithm to compute the transitive closure using saturation. Experimental results indicate that our improved algorithms achieve a clear speedup over previous algorithms in some cases. With the help of the new transitive closure computation algorithm, up to 10(exp 150) SCCs can be explored within a few seconds.

Zhao, Yang↗

Locality-Aware Scheduling for Scalable Heterogeneous Environments

Heterogeneous computing promise boost performance of scientific applications by allowing massively parallel execution of computational tasks. However, manually managing extremely heterogeneous, multi-device systems is complicated and may result in sub-optimal performance. Specifically, data management is an extremely challenging problem on multi-device systems. In this work, we introduce two locality-aware schedulers for the Minos Computing Library (MCL), an asynchronous, task-based programming model and runtime for extremely heterogeneous systems. The first scheduler implements a pure locality-aware algorithm to maximize data reuse, though it might incur in ”hot-spots” that limit system utilization. The second scheduler mitigates this drawback by dynamically targeting between locality-awareness and system utilization based on the current workload and available computing devices. Our results show that locality-awareness greatly benefit applications that exhibit data reuse, providing up to 6.9x and 7.9x over the original MCL scheduler and equivalent OpenCL implementations, respectively. Moreover, our schedulers introduce negligible overhead compared with the original MCL scheduler and achieve similar performance for applications that don’t benefit from data locality.

Architecture, co-design, Task-based programming mo↗

Collaborative Project: Building improved optimized parameter estimation algorithms to improve methane and nitrogen fluxes in a climate model

Soils in natural and managed ecosystems and wetlands are well known sources of methane, nitrous oxides, and reactive nitrogen gases, but the magnitudes of gas flux to the atmosphere are still poorly constrained. Thus, the reasons for the large increases in atmospheric concentrations of methane and nitrous oxide since the preindustrial time period are not well understood. The low atmospheric concentrations of methane and nitrous oxide, despite being more potent greenhouse gases than carbon dioxide, complicate empirical studies to provide explanations. In addition to climate concerns, the emissions of reactive nitrogen gases from soils are important to the changing nitrogen balance in the earth system, subject to human management, and may change substantially in the future. Thus improved modeling of the emission fluxes of these species from the land surface is important. Currently, there are emission modules for methane and some nitrogen species in the Community Earth System Model’s Community Land Model (CLM-ME/N); however, there are large uncertainties and problems in the simulations, resulting in coarse estimates. In this proposal, we seek to improve these emission modules by combining state-of-the-art process modules for emissions, available data, and new optimization methods. In earth science problems, we often have substantial data and knowledge of processes in disparate systems, and thus we need to combine data and a general process level understanding into a model for projections of future climate that are as accurate as possible. The best methodologies for optimization of parameters in earth system models are still being developed. In this proposal we will develop and apply surrogate algorithms that a) were especially developed for computationally expensive simulations like CLM-ME/N models; b) were (in the earlier surrogate optimization Stochastic RBF) demonstrated to perform very well on computationally expensive complex partial differential equations in earth science with limited numbers of simulations; and, c) will be (as part of the proposed research) significantly improved both by adding asynchronous parallelism, early truncation of unsuccessful simulations, and the improvement of both serial and parallel performance by the use of derivative and sensitivity information from global and local surrogate approximations S(x) . The algorithm development and testing will be focused on the CLM-ME/N model application, but the methods are general and are expected to also perform well on optimization for parameter estimation of other climate models and other classes of continuous multimodal optimization problems arising from complex simulation models. In addition, this proposal will compile available datasets of emissions of methane, nitrous oxides and reactive nitrogen species and develop protocols for site level comparisons with the CLM-ME/N. Once the model parameters are optimized against site level data, the model will be simulated at the global level and compared to atmospheric concentration measurements for the current climate, and future emissions will be estimated using climate change as simulated by the CESM. This proposal combines experts in earth system modeling, optimization, computer science, and process level understanding of soil gas emissions in an interdisciplinary team in order to improve the modeling of methane and nitrogen gas emissions. This proposal thus meets the requirements of the SciDAC RFP, by integrating state-of-the-art computer science and earth system to build an improved earth system model.

54 ENVIRONMENTAL SCIENCES↗

Addressing Global Data Dependencies in Heterogeneous Asynchronous Runtime Systems on GPUs

Large-scale parallel applications with complex global data dependencies beyond those of reductions pose significant scalability challenges in an asynchronous runtime system. Internodal challenges include identifying the all-to-all communication of data dependencies among the nodes. Intranodal challenges include gathering together these data dependencies into usable data objects while avoiding data duplication. This paper addresses these challenges within the context of a large-scale, industrial coal boiler simulation using the Uintah asynchronous many-task runtime system on GPU architectures. We show significant reduction in time spent analyzing data dependencies through refinements in our dependency search algorithm. Multiple task graphs are used to eliminate subsequent analysis when task graphs change in predictable and repeatable ways. Using a combined data store and task scheduler redesign reduces data dependency duplication ensuring that problems fit within host and GPU memory. Furthermore, these modifications did not require any changes to application code or sweeping changes to the Uintah runtime system. We report results running on the DOE Titan system on 119K CPU cores and 7.5K GPUs simultaneously. Our solutions can be generalized to other task dependency problems with global dependencies among thousands of nodes which must be processed efficiently at large scale.

97 MATHEMATICS AND COMPUTING↗

Fast GPU-Based Generation of Large Graph Networks From Degree Distributions

Synthetically generated, large graph networks serve as useful proxies to real-world networks for many graph-based applications. The ability to generate such networks helps overcome several limitations of real-world networks regarding their number, availability, and access. Here, we present the design, implementation, and performance study of a novel network generator that can produce very large graph networks conforming to any desired degree distribution. The generator is designed and implemented for efficient execution on modern graphics processing units (GPUs). Given an array of desired vertex degrees and number of vertices for each desired degree, our algorithm generates the edges of a random graph that satisfies the input degree distribution. Multiple runtime variants are implemented and tested: 1) a uniform static work assignment using a fixed thread launch scheme, 2) a load-balanced static work assignment also with fixed thread launch but with cost-aware task-to-thread mapping, and 3) a dynamic scheme with multiple GPU kernels asynchronously launched from the CPU. The generation is tested on a range of popular networks such as Twitter and Facebook, representing different scales and skews in degree distributions. Results show that, using our algorithm on a single modern GPU (NVIDIA Volta V100), it is possible to generate large-scale graph networks at rates exceeding 50 billion edges per second for a 69 billion-edge network. GPU profiling confirms high utilization and low branching divergence of our implementation from small to large network sizes. For networks with scattered distributions, we provide a coarsening method that further increases the GPU-based generation speed by up to a factor of 4 on tested input networks with over 45 billion edges.

97 MATHEMATICS AND COMPUTING↗

CGI delay compensation

Computer-generated graphics in real-time helicopter simulation produces objectionable scene-presentation time delays. In the flight simulation laboratory at Ames Research Center, it has been determined that these delays have an adverse influence on pilot performance during agressive tasks such as nap of the earth (NOE) maneuvers. Using contemporary equipment, computer generated image (CGI) time delays are an unavoidable consequence of the operations required for scene generation. However, providing that magnitude distortions at higher frequencies are tolerable, delay compensation is possible over a restricted frequency range. This range, assumed to have an upper limit of perhaps 10 or 15 rad/sec, conforms approximately to the bandwidth associated with helicopter handling qualities research. A compensation algorithm is introduced here and evaluated in terms of tradeoffs in frequency responses. The algorithm has a discrete basis and accommodates both a large, constant transport delay interval and a periodic delay interval, as associated with asynchronous operations.

Mcfarland, R. E.↗

CGI delay compensation

Computer-generated graphics in real-time helicopter simulation produces objectionable scene-presentation time delays. In the flight simulation laboratory at Ames Research Center, it has been determined that these delays have an adverse influence on pilot performance during aggressive tasks such as nap-of-the-earth (NOE) maneuvers. Using contemporary equipment, computer-generated image (CGI) time delays are an unavoidable consequence of the operations required for scene generation. However, providing that magnitide distortions at higher frequencies are tolerable, delay compensation is possible over a restricted frequency range. This range, assumed to have an upper limit of perhaps 10 or 15 rad/sec, conforms approximately to the bandwidth associated with helicopter handling qualities research. A compensation algorithm is introduced here and evaluated in terms of tradeoffs in frequency responses. The algorithm has a discrete basis and accommodates both a large, constant transport delay interval and a periodic delay interval, as associated with asynchronous operations.

Mcfarland, Richard E.↗

Asynchronous τ-leaping

Stochastic simulation of cell signaling pathways and genetic regulatory networks has contributed to the understanding of cell function; however, investigation of larger, more complicated systems requires computationally efficient algorithms. τ-leaping methods, which improve efficiency when some molecules have high copy numbers, either use a fixed leap size, which does not adapt to changing state, or recalculate leap size at a heavy computational cost. We present a hybrid simulation method for reaction-diffusion systems which combines exact stochastic simulation and τ-leaping in a dynamic way. Putative times of events are stored in a priority queue, which reduces the cost of each step of the simulation. For every reaction and diffusion channel at each step of the simulation the more efficient of an exact stochastic event or a τ-leap is chosen. This new approach removes the inherent trade-off between speed and accuracy in stiff systems which was present in all τ-leaping methods by allowing each reaction channel to proceed at its own pace. Both directions of reversible reactions and diffusion are combined in a single event, allowing bigger leaps to be taken. This improves efficiency for systems near equilibrium where forward and backward events are approximately equally frequent. Comparison with existing algorithms and behaviour for five test cases of varying complexity shows that the new method is almost as accurate as exact stochastic simulation, scales well for large systems, and for various problems can be significantly faster than τ-leaping.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Devastator Parallel Discrete Event Simulation Runtime (Devastator) v1.0

The Devastator runtime is a modern C++ implementation of optimistic parallel discrete event simulation methods. Devastator allows simulation application code to productively specify their component and event functionality with C++14 constructs. It utilizes GASNet-EX for distributed memory communication and includes parallel performance optimizations such as light-weight thread message queues and asynchronous GVT. Furthermore, it supports efficient event broadcasts and pause-rewind-resume functionality to support periodic load balancing and outer loop optimization algorithms.

Chan, Cy↗

Framework for Extensible, Asynchronous Task Scheduling (FEATS) in Fortran

Most parallel scientific programs contain compiler directives (pragmas) such as those from OpenMP, explicit calls to runtime library procedures such as those implementing the Message Passing Interface (MPI), or compiler-specific language extensions such as those provided by CUDA. By contrast, the recent Fortran standards empower developers to express parallel algorithms without directly referencing lower-level parallel programming models. Fortran’s parallel features place the language within the Partitioned Global Address Space (PGAS) class of programming models. When writing programs that exploit data-parallelism, application developers often find it straightforward to develop custom parallel algorithms. Problems involving complex, heterogeneous, staged calculations, however, pose much greater challenges. Such applications require careful coordination of tasks in a manner that respects dependencies prescribed by a directed acyclic graph. When rolling one’s own solution proves difficult, extending a customizable framework becomes attractive. The paper presents the design, implementation, and use of the Framework for Extensible Asynchronous Task Scheduling (FEATS), which we believe to be the first task-scheduling tool written in modern Fortran. We describe the benefits and compromises associated with choosing Fortran as the implementation language, and we propose ways in which future Fortran standards can best support the use case in this paper.

Richardson, Brad↗

Framework for Extensible, Asynchronous Task Scheduling (FEATS) in Fortran

Most parallel scientific programs contain compiler directives (pragmas) such as those from OpenMP, explicit calls to runtime library procedures such as those implementing the Message Passing Interface (MPI), or compiler-specific language extensions such as those provided by CUDA. By contrast, the recent Fortran standards empower developers to express parallel algorithms without directly referencing lower-level parallel programming models. Fortran’s parallel features place the language within the Partitioned Global Address Space (PGAS) class of programming models. When writing programs that exploit data-parallelism, application developers often find it straightforward to develop custom parallel algorithms. Problems involving complex, heterogeneous, staged calculations, however, pose much greater challenges. Such applications require careful coordination of tasks in a manner that respects dependencies prescribed by a directed acyclic graph. When rolling one’s own solution proves difficult, extending a customizable framework becomes attractive. The paper presents the design, implementation, and use of the Framework for Extensible Asynchronous Task Scheduling (FEATS), which we believe to be the first task-scheduling tool written in modern Fortran. We describe the benefits and compromises associated with choosing Fortran as the implementation language, and we propose ways in which future Fortran standards can best support the use case in this paper.

Modern Fortran↗

Distributed approximate minimal Steiner trees with millions of seed vertices on billion-edge graphs

In this report, we present a parallel 2-approximation Steiner minimal tree algorithm and its MPI-based distributed implementation. In place of expensive distance computations between all pairs of seed vertices, the solution we employ exploits a cheaper Voronoi cell computation. Our design leverages asynchronous processing and message prioritization to accelerate convergence of distance computations, and harnesses vertex and edge centric processing to offer fast time-to-solution. We demonstrate scalability and performance using real-world graphs with up to 128 billion edges and 512 compute nodes, and show the ability to find Steiner trees with up to one million seed vertices. Using 12 data instances, we present comparison with the state-of-the-art exact solver, SCIP-Jack, and two sequential 2-approximate algorithms. We empirically show that, on average, the total distance of the Steiner tree identified by our solution is 1.1290 times greater than the Steiner minimal tree – well within the theoretical approximation bound of 2.

97 MATHEMATICS AND COMPUTING↗