Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel computer architecture”

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 955 records · Page 53

MILC staggered conjugate gradient performance on Intel KNL

We review our work done to optimize the staggered conjugate gradient (CG) algorithm in the MILC code for use with the Intel Knights Landing (KNL) architecture. KNL is the second gener- ation Intel Xeon Phi processor. It is capable of massive thread parallelism, data parallelism, and high on-board memory bandwidth and is being adopted in supercomputing centers for scientific research. The CG solver consumes the majority of time in production running, so we have spent most of our effort on it. We compare performance of an MPI+OpenMP baseline version of the MILC code with a version incorporating the QPhiX staggered CG solver, for both one-node and multi-node runs.

97 MATHEMATICS AND COMPUTING↗

Implementing a neural network interatomic model with performance portability for emerging exascale architectures

The two main thrusts of computational science are increasingly accurate predictions and faster calculations; to this end, the zeitgeist in molecular dynamics (MD) simulations is pursuing machine learned and data driven interatomic models, e.g. neural network potentials, and novel hardware architectures, e.g. GPUs. Current implementations of neural network potentials are orders of magnitude slower than traditional interatomic models and while looming exascale computing offers the ability to run large, accurate simulations with these models, achieving portable performance for MD with new and varied exascale hardware requires rethinking traditional algorithms, using novel data structures, and library solutions. We re-implement a neural network interatomic model in CabanaMD, an MD proxy application, built on libraries developed for performance portability. Our implementation shows significantly improved thread scaling in this complex kernel as compared to a current LAMMPS implementation, across both strong and weak scaling. Our single-source solution enables simulations up to 20 million atoms on a single CPU node and 4 million atoms with improved performance on a single GPU. Furthermore, we also explore parallelism and data layout choices (using flexible data structures called AoSoAs) and their effect on performance, seeing up to ~50% and ~5% improvements in performance on a GPU by choosing the right level of parallelism and data layout respectively.

97 MATHEMATICS AND COMPUTING↗

Revisiting Huffman Coding: Toward Extreme Performance on Modern GPU Architectures

Today's high-performance computing (HPC) applications are producing vast volumes of data, which are challenging to store and transfer efficiently during the execution, such that data compression is becoming a critical technique to mitigate the storage burden and data movement cost. Huffman coding is arguably the most efficient Entropy coding algorithm in information theory, such that it could be found as a fundamental step in many modern compression algorithms such as DEFLATE. On the other hand, today's HPC applications are more and more relying on the accelerators such as GPU on supercomputers, while Huffman encoding suffers from low throughput on GPUs, resulting in a significant bottleneck in the entire data processing. In this paper, we propose and implement an efficient Huffman encoding approach based on modern GPU architectures, which addresses two key challenges: (1) how to parallelize the entire Huffman encoding algorithm, including codebook construction, and (2) how to fully utilize the high memory-bandwidth feature of modern GPU architectures. The detailed contribution is fourfold. (1) We develop an efficient parallel codebook construction on GPUs that scales effectively with the number of input symbols. (2) We propose a novel reduction based encoding scheme that can efficiently merge the codewords on GPUs. (3) We optimize the overall GPU performance by leveraging the state-of-the-art CUDA APIs such as Cooperative Groups. (4) We evaluate our Huffman encoder thoroughly using six real-world application datasets on two advanced GPUs and compare with our implemented multi-threaded Huffman encoder. Experiments show that our solution can improve the encoding throughput by up to 5.0x and 6.8x on NVIDIA RTX 5000 and V100, respectively, over the state-of-the-art GPU Huffman encoder, and by up to 3.3x over the multi-thread encoder on two 28-core Xeon Platinum 8280 CPUs.

Tian, Jiannan↗

The New MCNP6 Depletion Capability

The first MCNP based inline Monte Carlo depletion capability was officially released from the Radiation Safety Information and Computational Center as MCNPX 2.6.0. Both the MCNP5 and MCNPX codes have historically provided a successful combinatorial geometry based, continuous energy, Monte Carlo radiation transport solution for advanced reactor modeling and simulation. However, due to separate development pathways, useful simulation capabilities were dispersed between both codes and not unified in a single technology. MCNP6, the next evolution in the MCNP suite of codes, now combines the capability of both simulation tools, as well as providing new advanced technology, in a single radiation transport code. We describe here the new capabilities of the MCNP6 depletion code dating from the official RSICC release MCNPX 2.6.0, reported previously, to the now current state of MCNP6. NEA/OECD benchmark results are also reported. The MCNP6 depletion capability enhancements beyond MCNPX 2.6.0 reported here include: (1) new performance enhancing parallel architecture that implements both shared and distributed memory constructs; (2) enhanced memory management that maximizes calculation fidelity; and (3) improved burnup physics for better nuclide prediction. MCNP6 depletion enables complete, relatively easy-to-use depletion calculations in a single Monte Carlo code. The enhancements described here help provide a powerful capability as well as dictate a path forward for future development to improve the usefulness of the technology.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Revisiting Huffman Coding: Toward Extreme Performance on Modern GPU Architectures

Today's high-performance computing (HPC) applications are producing vast volumes of data, which are challenging to store and transfer efficiently during the execution, such that data compression is becoming a critical technique to mitigate the storage burden and data movement cost. Huffman coding is arguably the most efficient Entropy coding algorithm in information theory, such that it could be found as a fundamental step in many modern compression algorithms such as DEFLATE. On the other hand, today's HPC applications are more and more relying on the accelerators such as GPU on supercomputers, while Huffman encoding suffers from low throughput on GPUs, resulting in a significant bottleneck in the entire data processing. In this paper, we propose and implement an efficient Huffman encoding approach based on modern GPU architectures, which addresses two key challenges: (1) how to parallelize the entire Huffman encoding algorithm, including codebook construction, and (2) how to fully utilize the high memory-bandwidth feature of modern GPU architectures. The detailed contribution is fourfold. (1) We develop an efficient parallel codebook construction on GPUs that scales effectively with the number of input symbols. (2) We propose a novel reduction based encoding scheme that can efficiently merge the codewords on GPUs. (3) We optimize the overall GPU performance by leveraging the state-of-the-art CUDA APIs such as Cooperative Groups. (4) We evaluate our Huffman encoder thoroughly using six real-world application datasets on two advanced GPUs and compare with our implemented multithreaded Huffman encoder. Experiments show that our solution can improve the encoding throughput by up to 5.0× and 6.8× on NVIDIA RTX 5000 and V100, respectively, over the state-of-the-art GPU Huffman encoder, and by up to 3.3× over the multithread encoder on two 28-core Xeon Platinum 8280 CPUs.

Tian, Jiannan↗

The Evolution of the Internet Community and the"Yet-to-Evolve" Smart Grid Community: Parallels and Lessons-to-be-Learned

The Smart Grid envisions a transformed US power distribution grid that enables communicating devices, under human supervision, to moderate loads and increase overall system stability and security. This vision explicitly promotes increased participation from a community that, in the past, has had little involvement in power grid operations -the consumer. The potential size of this new community and its member's extensive experience with the public Internet prompts an analysis of the evolution and current state of the Internet as a predictor for best practices in the architectural design of certain portions of the Smart Grid network. Although still evolving, the vision of the Smart Grid is that of a community of communicating and cooperating energy related devices that can be directed to route power and modulate loads in pursuit of an integrated, efficient and secure electrical power grid. The remaking of the present power grid into the Smart Grid is considered as fundamentally transformative as previous developments such as modern computing technology and high bandwidth data communications. However, unlike these earlier developments, which relied on the discovery of critical new technologies (e.g. the transistor or optical fiber transmission lines), the technologies required for the Smart Grid currently exist and, in many cases, are already widely deployed. In contrast to other examples of technical transformations, the path (and success) of the Smart Grid will be determined not by its technology, but by its system architecture. Fortunately, we have a recent example of a transformative force of similar scope that shares a fundamental dependence on our existing communications infrastructure - namely, the Internet. We will explore several ways in which the scale of the Internet and expectations of its users have shaped the present Internet environment. As the presence of consumers within the Smart Grid increases, some experiences from the early growth of the Internet are expected to be informative and pertinent.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Memory-Aware External Facelist Calculation: A Data-Parallel Atomic Hash Counting Approach

Unstructured volumetric meshes serve as fundamental data representations in various scientific simulations and analyses. They play a crucial role in representing complex computational domains and are essential for important numerical techniques, such as finite element analysis. Whenever such a mesh is read from a file, streamed in-situ, or generated by algorithms, scientific visualization libraries rely on calculating the external surface of a geometry, named “external facelist”, to produce a polygonal mesh for rendering. Consequently, external facelist calculation has become one of the most widely used algorithms in the scientific visualization domain, necessitating optimal performance. In this paper, we explore relevant work on external facelist calculation algorithms in two common visualization libraries, VTK and Viskores, assess their performance and memory constraints, and introduce a novel memory-aware external facelist calculation algorithm employing an atomic hash counting approach. This algorithm fully leverages Viskores' data-parallel primitive operations, facilitating its execution across diverse many-core architectures. Our algorithm features the lowest memory footprint on the GPU and the second-lowest on the CPU among all evaluated methods, and it also delivers the fastest performance on both CPU and GPU. It has been made available under an open-source license in the VTK and Viskores visualization systems.

Tsalikis, Spiros [Kitware] (ORCID:0000000151137195↗

Parametric Multi-Level Tiling of Imperfectly Nested Loops

Tiling is a critical loop transformation for generating high-performance code on modern architectures. Efficient generation of multilevel tiled code is essential to exploit several levels of parallelism and/or to maximize data reuse in deep memory hierarchies. Tiled loops with parameterized tile sizes (not compile time constants) facilitate runtime feedback and dynamic optimizations used in iterative compilation and automatic tuning. The existing parametric multilevel tiling approach has focused on transformation for perfectly nested loops, where all assignment statements are contained inside the innermost loop of a loop nest. Previous solutions to tiling for imperfect loop nests are limited to the case where tile sizes are fixed. In this paper, we present an approach to parameterized multilevel tiling for imperfectly nested loops. Our tiling algorithm generates loops that iterate over full rectangular tiles that are amenable for potential compiler optimizations such as register tiling. Experimental results using a number of computational benchmarks demonstrate the effectiveness of our tiling approach.

97 MATHEMATICS AND COMPUTING↗

Direct Numerical Simulation of Hypersonic Turbulent Boundary Layer Flow using SPARC: Initial Evaluation

This report documents the initial testing of the Sandia Parallel Aerodynamics and Reentry Code (SPARC) to directly simulate hypersonic, turbulent boundary layer flow over a sharp 7- degree half-angle cone. This type of computation involves a tremendously large range of scales both in time and space, requiring a large number of grid cells and the efficient utilization of a large pool of resources. The goal of the simulation is to mimic and verify a wind tunnel experiment that seeks to measure the turbulent surface pressure fluctuations. These data are necessary for building a model to predict random vibration loading in the reentry flight environment. A low-dissipation flux scheme in SPARC is used on a 2.7 billion cell mesh to capture the turbulent fluctuations in the boundary layer flow. The grid is divided into 115200 partitions and simulated using the Knight's Landings (KNL) partition of the Trinity system. The parallel performance of SPARC is explored on the Trinity system, as well as some of the other new architectures. Extracting data from the simulation shows good agreement with the experiment as well as a colleague's simulation. The data provide a guide for which a new model can be built for better prediction of the reentry random vibration loads.

42 ENGINEERING↗

KokkACC: Enhancing Kokkos with OpenACC

Template metaprogramming is gaining popularity as a high-level solution for achieving performance portability on heterogeneous computing resources. Kokkos is a representative approach that offers programmers high-level abstractions for generic programming while most of the device-specific code generation and optimizations are delegated to the compiler through template specializations. For this, Kokkos provides a set of device-specific code specializations in multiple back ends, such as CUDA and HIP. Unlike CUDA or HIP, OpenACC is a high-level and directive-based programming model. This descriptive model allows developers to insert hints (pragmas) into their code that help the compiler to parallelize the code. The compiler is responsible for the transformation of the code, which is completely transparent to the programmer. This paper presents an OpenACC back end for Kokkos: KokkACC. As an alternative to Kokkos’s existing device-specific back ends, KokkACC is a multi-architecture back end providing a high-productivity programming environment enabled by OpenACC’s high-level and descriptive programming model. Moreover, we have observed competitive performance; in some cases, KokkACC is faster (up to 9×) than NVIDIA’s CUDA back end and much faster than OpenMP’s GPU offloading back end. This work also includes implementation details and a detailed performance study conducted with a set of mini-benchmarks (AXPY and DOT product) and three mini-apps (LULESH, miniFE and SNAP, a LAMMPS proxy mini-app).

Valero Lara, Pedro↗

Introduction: Neuromorphic Materials

The explosive growth in data collection and the need to process it efficiently, as well as the desire to automate increasingly complex tasks in transportation, medical care, manufacturing, security and many other fields have motivated a growing interest in neuromorphic computing. Unlike the binary, transistorbased ON/OFF logic gates and separate logic and memory functionalities employed in digital computing, neuromorphic computing is inspired by animal brains that use interconnected synapses and neurons to perform processing, storage and transmission of information at the same location, while only consuming ~20 W or less of power. Motivated by the brain’s efficiency, adaptability, self-learning and resiliency qualities, neuromorphic computing can be broadly defined as an approach to processing and storing information using hardware and algorithms inspired by models of biological neural systems. Present research in neuromorphic computing encompasses approaches that vary significantly in their degree of neuro-inspiration, from systems that only incorporate features such as asynchronous, event-driven operation or use crossbar arrays of non-volatile memory (NVM) elements to accelerate deep neural networks (DNNs), to designs that embrace the extreme parallelism, sparsity, reconfigurability, adaptability, complexity and stochasticity observed in nervous systems. The term ‘neuromorphic’ computing is often credited to Carver Mead, who in the 1980s investigated Si-based analog electronics to replicate functions of the animal retina. Earlier important advances in this field include the work of Frank Rosenblatt, who proposed the concept of the perceptron, Bernard Widrow, who used this concept to build one of the first analog neural networks, the Adaline and many other researchers (see ref. 6 for an historical perspective on neuromorphic computing). With the recent increase in the use of artificial intelligence and large language models, and rising concerns over the associated energy costs, interest in neuromorphic hardware has expanded rapidly. According to some estimates, driven largely by the drastic growth in the training use of artificial intelligence (AI) models using the current computing architectures, the energy cost of computing is projected to reach the energy supply worldwide by 2045. Furthermore, while this is not a realistic outcome, it means that, if more efficient computing technologies are not developed -- soon -- the world will soon become one where demand for energy and market constraints limit the continued increase of societal access to AI and cloud services from data centers. Data centers used for training and use of these models consume hundreds of terawatt hours of electricity, already past 4% of the US electricity demand.

Circuits↗

A massively parallel and scalable multi-CPU material point method

Harnessing the power of modern multi-GPU architectures, we present a massively parallel simulation system based on the Material Point Method (MPM) for simulating physical behaviors of materials undergoing complex topological changes, self-collision, and large deformations. Our system makes three critical contributions. First, we introduce a new particle data structure that promotes coalesced memory access patterns on the GPU and eliminates the need for complex atomic operations on the memory hierarchy when writing particle data to the grid. Second, we propose a kernel fusion approach using a new Grid-to-Particles-to-Grid (G2P2G) scheme, which efficiently reduces GPU kernel launches, improves latency, and significantly reduces the amount of global memory needed to store particle data. Finally, we introduce optimized algorithmic designs that allow for efficient sparse grids in a shared memory context, enabling us to best utilize modern multi-GPU computational platforms for hybrid Lagrangian-Eulerian computational patterns. We demonstrate the effectiveness of our method with extensive benchmarks, evaluations, and dynamic simulations with elastoplasticity, granular media, and fluid dynamics. In comparisons against an open-source and heavily optimized CPU-based MPM codebase [Fang et al. 2019] on an elastic sphere colliding scene with particle counts ranging from 5 to 40 million, our GPU MPM achieves over 100x per-time-step speedup on a workstation with an Intel 8086K CPU and a single Quadro P6000 GPU, exposing exciting possibilities for future MPM simulations in computer graphics and computational science. Moreover, compared to the state-of-the-art GPU MPM method [Hu et al. 2019a], we not only achieve 2x acceleration on a single GPU but our kernel fusion strategy and Array-of-Structs-of-Array (AoSoA) data structure design also generalizes to multi-GPU systems. Our multi-GPU MPM exhibits near-perfect weak and strong scaling with 4 GPUs, enabling performant and large-scale simulations on a 10243 grid with close to 100 million particles with less than 4 minutes per frame on a single 4-GPU workstation and 134 million particles with less than 1 minute per frame on an 8-GPU workstation.

Wang, Xinlei↗

Memory-efficient optimization of Gyrokinetic particle-to-grid interpolation for multicore processors

We present multicore parallelization strategies for the particle-to-grid interpolation step in the Gyrokinetic Toroidal Code (GTC), a 3D particle-in-cell (PIC) application to study turbulent transport in magnetic-confinement fusion devices. Particle-grid interpolation is a known performance bottleneck in several PIC applications. In GTC, this step involves particles depositing charges to a 3D toroidal mesh, and multiple particles may contribute to the charge at a grid point. We design new parallel algorithms for the GTC charge deposition kernel, and analyze their performance on three leading multicore platforms. We implement thirteen different variants for this kernel and identify the best-performing ones given typical PIC parameters such as the grid size, number of particles per cell, and the GTC-specific particle Larmor radius variation. We find that our best strategies can be 2x faster than the reference optimized MPI implementation, and our analysis provides insight into desirable architectural features for high-performance PIC simulation codes.

97 MATHEMATICS AND COMPUTING↗

CLARA: A Contemporary Approach to Physics Data Processing

In traditional physics data processing (PDP) systems, data location is static and is accessed by analysis applications. In comparison, CLARA (CLAS12 Reconstruction and Analysis framework) is an environment where data processing algorithms filter continuously flowing data. In CLARA's domain of loosely coupled services, data is not stored, but rather flows from one service to another, mutating constantly along the way. Agents, performing event processing, can then subscribe to particular data/events at any stage of the data transformation, and make intricate decisions (e.g. particle ID) by correlating events from multiple, parallel data streams and/or services. This paper presents a PDP application development framework based on service oriented and event driven architectures. This system allows users to design (Java, C++, and Python languages are supported) and deploy data processing services, as well as dynamically compose PDP applications using available services. The PDP service bus provides a layer on top of a distributed pub-sub middleware implementation, which allows complex service composition and integration without writing code. Examples of service creation and deployment, along with the CLAS12 track reconstruction application design will be presented.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Compiling MPI for Many-Core Systems

Processors with multiple (or many) cores and shared memory are becoming ubiquitous across the com puting spectrum. MPI, the current de facto program ming model for scalable parallel applications, enforces copies between source and target processes and thus can not fully utilize shared memory and cache architectures of modern machines. To enable MPI-based programs to more fully exploit features of multi- and many-core architectures, we present a compiler-based transformation that transforms MPI processes into threads and fuses message serialization and deserialization loops such that send and receive calls can be replaced by direct memory accesses. Our compiler replaces most of the MPI communication functions with direct load/store accesses and our runtime provides a threaded MPI implementation to handle the remaining functions.

97 MATHEMATICS AND COMPUTING↗

Failure analysis of nuclear transient-tested UN tristructural isotropic fuel particles in a 3D printed SiC matrix

Fully ceramic microencapsulated fuel elements containing UN tristructural isotropic (TRISO) fuel particles within a 3D printed SiC matrix were subjected to transient testing with varying energy depositions. Detailed post-irradiation examinations were performed, including leaching in hot HNO3 and post-leaching X-ray computed tomography, to quantify the percentage of failed TRISO particles and crack propagation within the particles and surrounding fuel matrix. In parallel, detailed finite element analyses were performed for comparison with experimental findings and to better evaluate transient failure modes. The lowest transient energy deposition—which still exceeded bounding values for high-temperature gas-cooled reactor applications—resulted in no detectable TRISO particle failures or matrix cracking, which was consistent with the simulations. Simulations of the higher-energy transients for which significant TRISO particle failure was expected were generally able to reproduce the transient temperatures and matrix cracking. Thus, the TRISO particle failures were explained based on the effects of local SiC matrix thickness and porosity. Results generally confirmed the high strength of the additively manufactured SiC matrix but also affirmed the need for a modified UN TRISO architecture to prevent SiC matrix cracks from propagating through TRISO layers. This unique failure mode has not historically been considered for TRISO fuels contained in weaker graphite matrices.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Fixing Amdahl's Law within the Limits of Accelerated Systems: FALLACY

Closeout report for FALLACY project. The performance of Data Model Convergence Initiative (DMC) applications on parallel machines is far below the limit set by Amdahl’s law. Whether the machine is based on many-core, GPUs, FPGAs, or a heterogeneous combination, usually the most significant bottleneck is accessing data from the memory system. Aligning with DMC’s HW/architecture thrust, this project developed a set of memory-centric tools called ‘MemGaze’ that inform the HW/SW stack about an application’s memory behavior, including data access latency and diagnosing poor data layout and data composition. Our approach uses architectural modeling and analysis of workload data accesses.

97 MATHEMATICS AND COMPUTING↗

Novel Solver Algorithms for Nearly Singular Linear Systems Arising in Combustion Modelling

Direct Numerical Simulations of realistic combustion devices are extremely challenging due to the wide separation of scales in the simulation, for example an internal combustion (IC) engine chamber, and the flame thickness of a high-pressure flame. The PeleLMeX solver uses adaptive mesh refinement (AMR) to evolve multi-species reacting flows in the low Mach number limit at the Exascale and relies on an embedded boundary (EB) approach to represent complex geometries. In that framework, the EB geometries often give rise to very small cut-cells along the boundary, which translate into extreme ill-conditioning of the pressure-projection, with eigenvalues that span 15-16 orders of magnitude. In this talk, we focus on the case of a typical IC piston bowl geometry for which we present on a novel approach towards solving these nearly singular linear systems with ILU-based, C-AMG smoothers on massively parallel architectures. In particular, we use scaling and equilibration algorithms to handle the non-normality of the upper triangular factors. This enables us to approximate the highly sequential triangular solve algorithm, embedded in the AMG smoothing-solve phase, with Jacobi iterations. This approximation can be written as a convergent Neumann series whose terms are composed of highly parallel sparse matrix vector multiplications. The result is an algorithm that substantially decreases setup and solve time, compared to state-of-the-art, for these challenging linear systems.

combustion modelling↗