Search NASA⌕ Search

SEARCH · Search NASA

Results for “Threading”

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 1,423 records · Page 79

Parallel k-means Clustering of Geospatial Data Sets Using Manycore CPU Architectures

The increasing availability of high-resolution geospatiotemporal data sets from sources such as observatory networks, remote sensing platforms, and computational Earth system models has opened new possibilities for knowledge discovery and mining of weather, climate, ecological, and other geoscientific data sets fused from disparate sources. Many of the standard tools used on individual workstations are impractical for the analysis and synthesis of data sets of this size; however, new algorithmic approaches that can effectively utilize the complex memory hierarchies and the extremely high levels of parallelism available in state-of-the-art high-performance computing platforms can enable such analysis. Here, we describe pKluster, an open-source tool we have developed for accelerated k-means clustering of geospatial and geospatiotemporal data, and discuss algorithmic modifications and code optimizations we have made to enable it to effectively use parallel machines based on novel CPU architectures—such as the Intel “Knights Landing” Xeon Phi and Skylake Xeon processors—with many cores and hardware threads, and employing significant single instruction, multiple data (SIMD) parallelism. We outline some applications of the code in ecology and climate science contexts and present a detailed discussion of the performance of the code for one such application, LiDAR-derived vertical vegetation structure classification.

Mills, Richard T↗

A Novel Multithreaded Algorithm For Extracting Maximal Chordal Subgraphs

Chordal graphs are triangulated graphs where any cycle larger than three is bisected by a chord. Many combinatorial optimization problems such as computing the minimum fill-in, the size of the maximum clique and the chromatic number are NP-hard on general graphs but have polynomial time solutions on chordal graphs. In this paper, we present a novel multithreaded algorithm to extract a maximal chordal subgraph from a general graph. Our algorithm is based on an iterative approach where each thread can asynchronously update a subset of edges that are dynamically assigned to it. We implement our algorithm on two different multithreaded architectures – Cray XMT, a massively multithreaded platform, and AMD Magny-Cours, a shared memory multicore platform. In addition to the proof of correctness, we present the performance of our algorithm using a testset of carefully generated synthetical graphs with up to half-a-billion edges and real world networks from gene correlation studies. We demonstrate that our algorithm achieves high scalability for all inputs on both types of architectures.

Halappanavar, Mahantesh↗

Argo NodeOS: Toward Unified Resource Management for Exascale

Exascale systems are expected to feature hundreds of thousands of compute nodes with hundreds of hardware threads and complex memory hierarchies with a mix of on-package and persistent memory modules. In this context, the Argo project is developing a new operating system for exascale machines. Targeting production workloads using workflows or coupled codes, we improve the Linux kernel on several fronts. We extend the memory management of Linux to be able to subdivide NUMA memory nodes, allowing better resource partitioning among processes running on the same node. We also add support for memory-mapped access to node- local, PCIe-attached NVRAM devices and introduce a new scheduling class targeted at parallel runtimes supporting user- level load balancing.

Perarnau, Swann↗

Towards Highly Scalable Ab Initio Molecular Dynamics (AIMD) Simulations on the Intel Knights Landing Manycore Processor

The Ab Initio Molecular Dynamics (AIMD) method allows scientists to treat the dynamics of molecular and condensed phase systems while retaining a first-principles-based description of their interactions. This extremely important method has tremendous computational requirements, because the electronic Schr¨odinger equation, approximated using Kohn-Sham Density Functional Theory (DFT), is solved at every time step. With the advent of manycore architectures, application developers have a significant amount of processing power within each compute node that can only be exploited through massive parallelism. A compute intensive application such as AIMD forms a good candidate to leverage this processing power. In this paper, we focus on adding thread level parallelism to the plane wave DFT methodology implemented in NWChem. Through a careful optimization of tall-skinny matrix products, which are at the heart of the Lagrange multiplier and nonlocal pseudopotential kernels, as well as 3D FFTs, our OpenMP implementation delivers excellent strong scaling on the latest Intel Knights Landing (KNL) processor. We assess the efficiency of our Lagrange multiplier kernels by building a Roofline model of the platform, and verify that our implementation is close to the roofline for various problem sizes. Finally, we present strong scaling results on the complete AIMD simulation for a 64 water molecules test case, that scales up to all 68 cores of the Knights Landing processor.

Environmental Molecular Sciences Laboratory↗

Localized Fault Recovery for Nested Fork-Join Programs

Nested fork-join programs scheduled using work stealing can automatically balance load and adapt to changes in the execution environment. In this paper, we design an approach to efficiently recover from faults encountered by these programs. Specifically, we focus on localized recovery of the task space in the presence of fail-stop failures. We present an approach to efficiently track, under work stealing, the relationships between the work executed by various threads. This information is used to identify and schedule the tasks to be re-executed without interfering with normal task execution. The algorithm precisely computes the work lost, incurs minimal re-execution overhead, and can recover from an arbitrary number of failures. Experimental evaluation demonstrates low overheads in the absence of failures, recovery overheads on the same order as the lost work, and much lower recovery costs than alternative strategies.

Kestor, Gokcen↗

Distributed Louvain Algorithm for Graph Community Detection

In most real-world networks, the nodes/vertices tend to be organized into tightly-knit modules known as communities or clusters, such that nodes within a community are more likely to be “related” to one another than they are to the rest of the network. The goodness of partitioning into communities is typically measured using a well known measure called modularity. However, modularity optimization is an NP-complete problem. In 2008, Blondel, et al. introduced a multi-phase, iterative heuristic for modularity optimization, called the Louvain method. Owing to its speed and ability to yield high quality communities, the Louvain method continues to be one of the most widely used tools for serial community detection. In this paper, we present the design of a distributed memory implementation of the Louvain algorithm for parallel community detection. Our approach begins with an arbitrarily partitioned distributed graph input, and employs several heuristics to speedup the computation of the different steps of the Louvain algorithm. We evaluate our implementation and its different variants using real-world networks from various application domains (including internet, biology, social networks). Our MPI+OpenMP implementation yields about 7x speedup (on 4K processes) for soc-friendster network (1.8B edges) over a state-of-the-art shared-memory multicore implementation (on 64 threads), without compromising output quality. Furthermore, our distributed implementation was able to process a larger graph (uk-2007; 3.3B edges) in 32 seconds on 1K cores (64 nodes) of NERSC Cori, when the state-of-the-art shared-memory implementation failed to run due to insufficient memory on a single Cori node containing 128 GB.

High performance computing, graph algorithms, Dist↗

Runtime Concurrency Control and Operation Scheduling for High Performance Neural Network Training

In this paper, we extend an existing runtime system (the TensorFlow runtime) to enable automatic concurrency control and scheduling of operations. We explore performance modeling to predict the performance of operations with various thread- level parallelism. Our performance model is highly accurate and lightweight. Leveraging the performance model, our runtime system employs a set of scheduling strategies that co-run opera- tions to improve hardware utilization and system throughput. Our runtime system demonstrates a significant performance benefit. Comparing with using the recommended configurations for concurrency control and operation scheduling in TensorFlow, our approach achieves 36% performance (execution time) im- provement on average (up to 49%) for four neural network models, and achieves high performance close to the optimal one manually obtained by the user.

Liu, Jiawen↗

LOGAN: High-Performance GPU-Based X-Drop Long-Read Alignment

Pairwise sequence alignment is one of the most computationally intensive kernels in genomic data analysis, accounting for more than 90% of the runtime for key bioinformatics applications. This method is particularly expensive for third-generation sequences due to the high computational cost of analyzing sequences of length between 1Kb and 1Mb. Given the quadratic overhead of exact pairwise algorithms for long alignments, the community primarily relies on approximate algorithms that search only for high-quality alignments and stop early when one is not found. In this work, we present the first GPU optimization of the popular X-drop alignment algorithm, that we named LOGAN. Results show that our high-performance multi-GPU implementation achieves up to 181.6 GCUPS and speed-ups up to 6.6× and 30.7× using 1 and 6 NVIDIA Tesla V100, respectively, over the state-of-the-art software running on two IBM Power9 processors using 168 CPU threads, with equivalent accuracy. We also demonstrate a 2.3× LOGAN speed-up versus ksw2, a state-of-art vectorized algorithm for sequence alignment implemented in minimap2, a long-read mapping software. Furthermore, to highlight the impact of our work on a real-world application, we couple LOGAN with a many-to-many long-read alignment software called BELLA, and demonstrate that our implementation improves the overall BELLA runtime by up to 10.6×. Finally, we adapt the Roofline model for LOGAN and demonstrate that our implementation is near optimal on the NVIDIA Tesla V100s.

97 MATHEMATICS AND COMPUTING↗

Communication-Avoiding and Memory-Constrained Sparse Matrix-Matrix Multiplication at Extreme Scale

Sparse matrix-matrix multiplication (SpGEMM) is a widely used kernel in various graph, scientific computing and machine learning algorithms. In this paper, we consider SpGEMMs performed on hundreds of thousands of processors generating trillions of nonzeros in the output matrix. Distributed SpGEMM at this extreme scale faces two key challenges: (1) high communication cost and (2) inadequate memory to generate the output. Furthermore, we address these challenges with an integrated communication-avoiding and memory-constrained SpGEMM algorithm that scales to 262,144 cores (more than 1 million hardware threads) and can multiply sparse matrices of any size as long as inputs and a fraction of output fit in the aggregated memory. As we go from 16,384 cores to 262,144 cores on a Cray XC40 supercomputer, the new SpGEMM algorithm runs 10x faster when multiplying large-scale protein-similarity matrices.

97 MATHEMATICS AND COMPUTING↗

Scalable, Multithreaded, Partially-in-Place Sorting

A recent trend in hardware development is producing computing systems that are stretching the number of cores and size of shared-memory beyond where most fundamental serial algorithms perform well. The expectation is that this trend will continue. So it makes sense to rethink our fundamental algorithms such as sorting. There are many situations where data that needs to be sorted will actually fit into the shared memory so applications could benefit from an efficient parallel sorting algorithm. When sorting large data (at least hundreds of Gigabytes) in a single shared memory, there are two factors that affect the algorithm choice. First, does the algorithm sort in-place? And second, does the algorithm scale well beyond tens of threads? Surprisingly, existing algorithms posses either one of these factors, but not both. We present an approach that gracefully degrades in performance as the amount of available working memory decreases relative to the size of the input.

HPC↗

Radiative Heat Transfer Calculation on 16384 GPUs Using a Reverse Monte Carlo Ray Tracing Approach with Adaptive Mesh Refinement

Modeling thermal radiation is computationally challenging in parallel due to its all-to-all physical and resulting computational connectivity, and is also the dominant mode of heat transfer in practical applications such as next-generation clean coal boilers, being modeled by the Uintah framework. However, a direct all-to-all treatment of radiation is prohibitively expensive on large computers systems whether homogeneous or heterogeneous. DOE Titan and the planned DOE Summit and Sierra machines are examples of current and emerging GPU-based heterogeneous systems where the increased processing capability of GPUs over CPUs exacerbates this problem. These systems require that computational frameworks like Uintah leverage an arbitrary number of on-node GPUs, while simultaneously utilizing thousands of GPUs within a single simulation. We show that radiative heat transfer problems can be made to scale within Uintah on heterogeneous systems through a combination of reverse Monte Carlo ray tracing (RMCRT) techniques combined with AMR, to reduce the amount of global communication. In particular, significant Uintah infrastructure changes, including a novel lock and contention-free, thread-scalable data structure for managing MPI communication requests and improved memory allocation strategies were necessary to achieve excellent strong scaling results to 16384 GPUs on Titan.

Computer Science↗

Multijunction Solar Cells With Graded Buffer Bragg Reflectors

Metamorphic solar cells can have optimal bandgap combinations through the use of compositionally graded buffers, where the lattice constant is slowly varied over several microns of growth. Bragg reflectors consist of several microns of alternating layers with refractive index contrast and provide a useful internal reflection to multijunction solar cells with optically thin subcells. In this work, we implement distributed Bragg reflectors within the compositionally graded buffers of inverted metamorphic solar cells to add functionality to the buffer. The reflectance of this AlGaInAs 'graded buffer Bragg reflector' is very similar to the reflectance of a similar AlGaAs Bragg reflector external to a buffer as well as the reflectance predicted by the transfer matrix model, indicating that the roughness of the buffer does not drastically reduce the reflection. Reflectance of 72%, 91%, and 98% is achieved in 2, 4, and 8 um buffers using AlGaInAs layers that alternate between 30% and 70% aluminum content. Using a 2 um graded buffer Bragg reflector, the 1.0-eV mismatched subcell of a GaAs/GaInAs tandem has a minor increase in threading dislocation density compared to a standard graded buffer and a small, 20 mV, loss in voltage. As the buffer is thickened, the voltage loss is recuperated and excellent subcell voltages are achieved, indicating that the Bragg reflector is not severely hindering dislocation glide. We demonstrate that the benefits of the graded buffer Bragg reflector for optically thin subcells and subcells containing quantum wells, and conclude that Bragg reflectors can effectively be implemented within graded buffers, adding functionality without adding cost.

14 SOLAR ENERGY↗

GaAs Solar Cells on Nanopatterned Si Substrates

Integrating III-Vs onto Si is a promising route toward tandem photovoltaics and cost mitigation of III-V substrates. While many III-V/Si photovoltaic integration approaches have been studied, epitaxial growth on Si allows for fewer processing steps compared to other approaches. However, current epitaxial pathways utilize expensive techniques, such as thermal cycle annealing or thick buffer layers to control defect densities, undermining the low-cost goal of integrating III-Vs with Si. Here, we present single-junction GaAs solar cells grown directly on Si using selective area growth as an alternative low-cost technique to control material quality with a much thinner buffer. We demonstrate a 10.4%-efficient GaAs device grown on a V-grooved Si substrate, which achieved an antiphase domain-free III-V/Si interface and a threading dislocation density of 2 x 107 cm-2, despite the lack of a graded buffer. We compare this growth on V-grooved Si to solar cells grown on polished Si with and without a patterned silica buffer layer, which demonstrate efficiencies of 6.5% and 6.8%, respectively; the polished Si was patterned using nanoimprint lithography, which is a low-cost patterning technique compatible with III-V selective area growth. Cracking is found to be a critical challenge that hinders solar cell performance and is exacerbated by the V-grooved Si surface topography. The results in this paper provide a promising pathway toward high-efficiency, low-cost III-V/Si tandem photovoltaics.

14 SOLAR ENERGY↗

Investigation of Rear-Emitter GaAs 0.75 P 0.25 Top Cells for Application to III–V/Si Tandem Photovoltaics

A rear-emitter (rear-junction) n-on-p+ device design was investigated for use in metamorphic monolithic III-V/Si tandem solar cells as an alternative to the traditional front-emitter (front-junction) n+-on-p design for potentially greater resistance to threading dislocation-induced performance degradation. A comparison of MOCVD-grown rearversus front-emitter 1.7-eV bandgap GaAs 0.75 P 0.25 top cell isotypes demonstrated as 30-mV advantage in WOC for the rear-emitter design. Here, this reduction in WOC was determined to be nearly equally caused by a reduction in junction recombination current as well as reduced reverse saturation current from improved quasi-neutral region transport. These results suggest that the rear-emitter design may indeed be a promising pathway for application to metamorphic cells, including III-V/Si tandems, where achieving maximum voltage output is often hindered by elevated dislocation densities. However, further optimization of short-circuit current collection is needed to overcome suboptimal collection probability profiles that could limit the efficacy of such structures.

14 SOLAR ENERGY↗

Bristol Ridge: A 28-nm $\times$ 86 Performance-Enhanced Microprocessor Through System Power Management

Power management techniques can be effective at extracting more performance and energy efficiency out of mature systems on chip (SoCs). For instance, the peak performance of microprocessors is often limited by worst case technology (Vmax), infrastructure (thermal/electrical), and microprocessor usage assumptions. Performance/watt of microprocessors also typically suffers from guard bands associated with the test and binning processes as well as worst case aging/lifetime degradation. Similarly, on multicore processors, shared voltage rails tend to limit the peak performance achievable in low thread count workloads. In this paper, we describe five power management techniques that maximize the per-part performance under the before-mentioned constraints. Using these techniques, we demonstrate a net performance increase of up to 15% depending on the application and TDP of the SoC, implemented on 'Bristol Ridge,' a 28-nm CMOS, dual-core x 86 accelerated processing unit.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Visual Analytics for Power Grid Contingency Analysis

Contingency analysis is the process of employing different measures to model scenarios, analyze them, and then derive the best response to remove the threats. This application paper focuses on a class of contingency analysis problems found in the power grid management system. A power grid is a geographically distributed interconnected transmission network that transmits and delivers electricity from generators to end users. The power grid contingency analysis problem is increasingly important because of both the growing size of the underlying raw data that need to be analyzed and the urgency to deliver working solutions in an aggressive timeframe. Failure to do so may bring significant financial, economic, and security impacts to all parties involved and the society at large. The paper presents a scalable visual analytics pipeline that transforms about 100 million contingency scenarios to a manageable size and form for grid operators to examine different scenarios and come up with preventive or mitigation strategies to address the problems in a predictive and timely manner. Great attention is given to the computational scalability, information scalability, visual scalability, and display scalability issues surrounding the data analytics pipeline. Most of the large-scale computation requirements of our work are conducted on a Cray XMT multi-threaded parallel computer. The paper demonstrates a number of examples using western North American power grid models and data.

contingency analysis↗

The Trinity Center of Excellence Co-Design Best Practices

Co-design between application developers and the vendors providing high-performance computing systems is essential to efficiently using current-generation HPC systems and shaping next-generation architectures. Application developers are faced with the challenge of effectively orchestrating many hardware threads and exploiting vector instructions, which can require code modifications. Will refactoring applications for current-generation systems apply to future-generation systems? Is there help available for this enormous task? Over the past 10 years, a partnership has evolved to address these issues. Centers of Excellence are mutually beneficial collaborations between application developers and vendors; vendors receive a better understanding of the needs of their users, and users receive a better understanding of the characteristics of the hardware, system software, and tools available to address their challenges. Lastly, this article describes the partnership between two vendors, Cray and Intel, and the application teams working together in the Trinity Center of Excellence.

97 MATHEMATICS AND COMPUTING↗

Accelerating DCA++ (Dynamical Cluster Approximation) Scientific Application on the Summit supercomputer

Optimizing scientific applications on today’s accelerator-based high performance computing systems can be challenging, especially when multiple GPUs and CPUs with heterogeneous memories and persistent non-volatile memories are present. An example is Summit, an accelerator-based system at the Oak Ridge Leadership Computing Facility (OLCF) that is rated as the world’s fastest supercomputer to-date. New strategies are thus needed to expose the parallelism in legacy applications, while being amenable to efficient mapping to the underlying architecture.In this paper we discuss our experiences and strategies to port a scientific application, DCA++, to Summit. DCA++ is a highperformance research application that solves quantum manybody problems with a cutting edge quantum cluster algorithm, the dynamical cluster approximation.Our strategies aim to synergize the strengths of the different programming models in the code. These include: (a) streamlining the interactions between the CPU threads and the GPUs, (b) implementing computing kernels on the GPUs and decreasing CPU-GPU memory transfers, (c) allowing asynchronous GPU communications, and (d) increasing compute intensity by combining linear algebraic operations.Full-scale production runs using all 4600 Summit nodes attained a peak performance of 73.5 PFLOPS with a mixed precision implementation.We observed a perfect strong and weak scaling for the quantum Monte Carlo solver in DCA++, while encountering about 2× input/output (I/O) and MPI communication overhead on the time-to-solution for the full machine run. Our hardware agnostic optimizations are designed to alleviate the communication and I/O challenges observed, while improving the compute intensity and obtaining optimal performance on a complex, hybrid architecture like Summit.

Balduzzi, Giovanni↗