Balancing conflicting requirements for grid and particle decomposition in continuum-Lagrangian solvers
Not Available
Sitaraman, Hariswaran↗
SEARCH · Search NASA
Results for “Parallelization Strategy”
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.
Numerical analyses of exponential time-differencing schemes for the solution of atmospheric models
In high-resolution numerical weather prediction models, fast-moving acoustic waves must be treated in a stable manner. Among the implicit–explicit (IMEX) schemes used for the solution of these models, horizontally explicit, vertically implicit (HEVI) methods show good stability and scalability on massively parallel machines. In this work, we present two classes of exponential time-differencing (ETD) methods for atmospheric models that use a HEVI splitting strategy, one being a three-stage method with the addition of artificial diffusion, the second based on a Strang splitting approach. Overall, the stability properties of the methods are analyzed and numerical examples are provided, which compare time-step restrictions and cost-to-accuracy ratios of the new methods with those for existing approaches.
54 ENVIRONMENTAL SCIENCES↗
Toward a Predictive Understanding of Earth’s Microbiomes to Address 21st Century Challenges
Microorganisms have shaped our planet and its inhabitants for over 3.5 billion years. Humankind has had a profound influence on the biosphere, manifested as global climate and land use changes, and extensive urbanization in response to a growing population. The challenges we face to supply food, energy, and clean water while maintaining and improving the health of our population and ecosystems are significant. Given the extensive influence of microorganisms across our biosphere, we propose that a coordinated, cross-disciplinary effort is required to understand, predict, and harness microbiome function. From the parallelization of gene function testing to precision manipulation of genes, communities, and model ecosystems and development of novel analytical and simulation approaches, we outline strategies to move microbiome research into an era of causality. These efforts will improve prediction of ecosystem response and enable the development of new, responsible, microbiome-based solutions to significant challenges of our time.
59 BASIC BIOLOGICAL SCIENCES↗
Parallel derivative-free optimization for simulation-based design of behind-the-meter energy systems
In this work, the integrated design and dispatch of behind-the-meter or distributed resources (e.g. stationary battery storage and solar PV generation) is considered. A simulation-based framework is employed, generating high-fidelity results with closed-loop predictive control at a fine resolution, at the expense of high computational cost (several minutes to a few hours per design point). To address this challenge, parallel derivative-free design methods are considered. Four methods are compared, including state-of-the-art surrogate-based methods (Radial-Basis Functions and Gaussian processes) and sampling strategies, an evolutionary-based method, and a simple sequential grid refinement method. As a case study, two types of design problem with increasing complexity are considered, namely, the design of behind-the-meter resources (three design variables) and the inclusion of grid capacity (four design variables). The second yields a constrained design problem for which violations can only be determined after solving the computationally expensive simulation. For the three-dimensional case, all methods present a good performance, achieving a solution within 1% of the optimum after the first iteration, with the sequential grid refinement exhibiting the fastest convergence and achieving the best final objective value. This indicates that the parallel evaluation of multiple sampling points may be more important than the choice of method for small decision spaces. For the four-dimensional constrained case, the Genetic Algorithm presents the best tradeoff between performance and computational effort, while the rough objective function terrain generated by constraint violation penalties reduces the performance of surrogate-based methods. Contour plots with flat regions indicate flexibility in the optimal design and highlight the importance of characterizing the solution space.
24 POWER TRANSMISSION AND DISTRIBUTION↗
Kokkos: Enabling manycore performance portability through polymorphic memory access patterns
The manycore revolution can be characterized by increasing thread counts, decreasing memory per thread, and diversity of continually evolving manycore architectures. High performance computing (HPC) applications and libraries must exploit increasingly finer levels of parallelism within their codes to sustain scalability on these devices. We found that a major obstacle to performance portability is the diverse and conflicting set of constraints on memory access patterns across devices. Contemporary portable programming models address manycore parallelism (e.g., OpenMP, OpenACC, OpenCL) but fail to address memory access patterns. The Kokkos C++ library enables applications and domain libraries to achieve performance portability on diverse manycore architectures by unifying abstractions for both fine-grain data parallelism and memory access patterns. In this paper we describe Kokkos’ abstractions, summarize its application programmer interface (API), present performance results for unit-test kernels and mini-applications, and outline an incremental strategy for migrating legacy C++ codes to Kokkos. Furthermore, the Kokkos library is under active research and development to incorporate capabilities from new generations of manycore architectures, and to address a growing list of applications and domain libraries.
97 MATHEMATICS AND COMPUTING↗
Self-consistent field theory simulations of polymers on arbitrary domains
We introduce a framework for simulating the mesoscale self-assembly of block copolymers in arbitrary confined geometries subject to Neumann boundary conditions. We employ a hybrid finite difference/volume approach to discretize the mean-field equations on an irregular domain represented implicitly by a level-set function. The numerical treatment of the Neumann boundary conditions is sharp, i.e. it avoids an artificial smearing in the irregular domain boundary. This strategy enables the study of self-assembly in confined domains and enables the computation of physically meaningful quantities at the domain interface. In addition, we employ adaptive grids encoded with Quad-/Oc-trees in parallel to automatically refine the grid where the statistical fields vary rapidly as well as at the boundary of the confined domain. This approach results in a significant reduction in the number of degrees of freedom and makes the simulations in arbitrary domains using effective boundary conditions computationally efficient in terms of both speed and memory requirement. Finally, in the case of regular periodic domains, where pseudo-spectral approaches are superior to finite differences in terms of CPU time and accuracy, we use the adaptive strategy to store chain propagators, reducing the memory footprint without loss of accuracy in computed physical observables.
75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
Magic Pools: Parallel Assessment of Transposon Delivery Vectors in Bacteria
Transposon mutagenesis coupled to next-generation sequencing (TnSeq) is a powerful approach for discovering the functions of bacterial genes. However, the development of a suitable TnSeq strategy for a given bacterium can be costly and time-consuming. To meet this challenge, we describe a part-based strategy for constructing libraries of hundreds of transposon delivery vectors, which we term “magic pools.” Within a magic pool, each transposon vector has a different combination of upstream sequences (promoters and ribosome binding sites) and antibiotic resistance markers as well as a random DNA barcode sequence, which allows the tracking of each vector during mutagenesis experiments. To identify an efficient vector for a given bacterium, we mutagenize it with a magic pool and sequence the resulting insertions; we then use this efficient vector to generate a large mutant library. We used the magic pool strategy to construct transposon mutant libraries in five genera of bacteria, including three genera of the phylum Bacteroidetes .
59 BASIC BIOLOGICAL SCIENCES↗
ConnectIt: a framework for static and incremental parallel graph connectivity algorithms
Connected components is a fundamental kernel in graph applications. The fastest existing multicore algorithms for solving graph connectivity are based on some form of edge sampling and/or linking and compressing trees. However, many combinations of these design choices have been left unexplored. In this paper, we design the ConnectIt framework, which provides different sampling strategies as well as various tree linking and compression schemes. ConnectIt enables us to obtain several hundred new variants of connectivity algorithms, most of which extend to computing spanning forest. In addition to static graphs, we also extend ConnectIt to support mixes of insertions and connectivity queries in the concurrent setting. We present an experimental evaluation of ConnectIt on a 72-core machine, which we believe is the most comprehensive evaluation of parallel connectivity algorithms to date. Compared to a collection of state-of-the-art static multicore algorithms, we obtain an average speedup of 12.4x (2.36x average speedup over the fastest existing implementation for each graph). Using ConnectIt, we are able to compute connectivity on the largest publicly-available graph (with over 3.5 billion vertices and 128 billion edges) in under 10 seconds using a 72-core machine, providing a 3.1x speedup over the fastest existing connectivity result for this graph, in any computational setting. For our incremental algorithms, we show that our algorithms can ingest graph updates at up to several billion edges per second. To guide the user in selecting the best variants in ConnectIt for different situations, we provide a detailed analysis of the different strategies. Finally, we show how the techniques in ConnectIt can be used to speed up two important graph applications: approximate minimum spanning forest and SCAN clustering.
Computer Science↗
A Decoupled Droop Control Strategy for Cascaded Multicell Inverter with Low-Frequency Modulation
Cascaded multicell inverter (CMI), featured by its merits of modularity and fault-tolerance, is suitable for medium-voltage applications without using bulky step-up transformers. However, little research has reported grid-forming control of the CMI for the application in the power grid with high penetrations of inverter-interfaced generation. To bridge this research gap, this paper proposes a novel droop control strategy for the CMI with a low-frequency modulation to offer inertia and reactive power support to the power grid. A decoupled Q/V droop control strategy is proposed to eliminate the coupling effect from the P/ω control loop caused by the phase-angle difference between the voltages of the CMI and the grid. The output of the Q/V control loop is set as the reference voltage for each cell of the CMI, which makes the CMI naturally a voltage source to provide a grid-forming capability. Finally, the proposed control strategy shows great active and reactive power regulation and sharing capability, which is validated by simulation on two 100 kW, 2.4 kV CMIs connected in parallel to a power grid.
14 SOLAR ENERGY↗
sycl-reduction benchmark v1.0.0
Some different SYCL implementations of the reduction operation. The parallel reduction kernel aggregates vector elements into a single result. It consists of four ways to implement the reduction operation using two different data management strategies, which we could not find elsewhere as a single package.
Dufek, Amanda↗
Performance Analysis and Simulaion of the Hydraulic Scram System in TREAT Reactor
The Transient Reactor Test Facility (TREAT) at Idaho National Laboratory (INL) serves a vital role in nuclear fuel safety research, enabling transient experiments that simulate reactivity excursions and accident scenarios. Central to these operations is the transient control rod drive system (TCRDS), which drives rapid motion of the transient control rods such that TREAT can simulate rapid power changes typical of reactor accidents. The reliability and performance of this system are critical for protecting both fuel specimens and reactor infrastructure. This study presents the initial phase of a two-year investigation into the dynamics and reliability of the TREAT hydraulic TCRDS. Conducted in collaboration with INL, the research employs a combined computational and experimental approach to analyze the system's response time, pressure transients, and potential failure modes. Emphasis is placed on understanding how fluid characteristics influence the TCRDS’s ability to achieve both rapid power changes and mechanical stability. The TRDS and the skid that powers it will be analyzed throughout this investigation. Computational modeling using computational fluid dynamics (CFD) will simulate the hydraulic response under varying conditions. In parallel, experimental testing planned at INL will validate these models and capture key performance metrics. This paper outlines the system design, analytical framework, and modeling strategies that form the foundation for later testing. Ultimately, this work aims to support improvements to the TCRDS’s design and reliability, contributing to the broader goal of enhancing nuclear fuel safety and sustaining TREAT’s mission as a premier nuclear fuel test facility.
21 - SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLAN↗
Applying Machine Learning to Understand Write Performance of Large-scale Parallel Filesystems
In high-performance computing (HPC), I/O performance prediction offers the potential to improve the efficiency of scientific computing. In particular, accurate prediction can make runtime estimates more precise, guide users toward optimal checkpoint strategies, and better inform facility provisioning and scheduling policies. HPC I/O performance is notoriously difficult to predict and model, however, in large part because of inherent variability and a lack of transparency in the behaviors of constituent storage system components. In this work we seek to advance the state of the art in HPC I/O performance prediction by (1) modeling the mean performance to address high variability, (2) deriving model features from write patterns, system architecture and system configurations, and (3) employing Lasso regression model to improve model accuracy. We demonstrate the efficacy of our approach by applying it to a crucial subset of common HPC I/O motifs, namely, file-per-process checkpoint write workloads. We conduct experiments on two distinct production HPC platforms—Titan at the Oak Ridge Leadership Computing Facility and Cetus at the Argonne Leadership Computing Facility—to train and evaluate our models. We find that we can attain ≤ 30% relative error for 92.79% and 99.64% of the samples in our test set on these platforms, respectively.
Xie, Bing↗
Effect of the scanning strategy on the formation of residual stresses in additively manufactured Ti-6Al-4V
During the laser-powder bed fusion (L-PBF) process, high laser intensities, short interaction times and highly localized heat input drive large thermal gradients that result in a state of high residual stresses. Generally, the residual stresses that develop during the L-PBF process can compromise the performance of the component. Up to now, the literature has indicated that the magnitude of the residual stresses can be affected by various process parameters. In this study, all process parameters such as laser power and speed are held fixed and the focus is solely on the effect of the laser scan strategy on the three-dimensional residual stress state of L-PBF metallic components. Four Ti-6Al-4V bridge shaped components were built using island and continuous scanning patterns parallel and offset 45° from the sample axes. High-energy X-ray diffraction was used to determine the residual stress field in each of the components. Two of them were re-measured after being partially removed from the build plate. The assumptions implicit in diffraction measurements of stress are reviewed and discussed in depth because the unique microstructure associated with L-PBF Ti-6Al-4V renders the validity of those assumptions uncertain. Specifically, additional data was collected and analyzed to evaluate the relationship between grain scale and macroscopic scale stresses. The observed residual stresses were large, ½ to ¾ of the yield strength, particularly the build direction stresses near the lateral edges of the bridges. Here in this work, the higher stresses were observed in the bridges built via the island scan strategies, chiefly near the edges of the parts.
36 MATERIALS SCIENCE↗
Evict on write, a management strategy for a prefetch unit and/or first level cache in a multiprocessor system with speculative execution
In a multiprocessor system with at least two levels of cache, a speculative thread may run on a core processor in parallel with other threads. When the thread seeks to do a write to main memory, this access is to be written through the first level cache to the second level cache. After the write though, the corresponding line is deleted from the first level cache and/or prefetch unit, so that any further accesses to the same location in main memory have to be retrieved from the second level cache. The second level cache keeps track of multiple versions of data, where more than one speculative thread is running in parallel, while the first level cache does not have any of the versions during speculation. A switch allows choosing between modes of operation of a speculation blind first level cache.
97 MATHEMATICS AND COMPUTING↗
A PIC Bootstrapping Strategy for Exascale CFD-DEM Simulation
The exascale computing era comes with the release of MFIX-Exa, a new code for studying gas-particle fluidization using CFD-DEM and PIC models on massively parallel, heterogeneous high-performance computing architectures. However, as compute resources continue to grow in scale and complexity, so too does the impetus to use them efficiently. Here, we propose a novel bootstrapping method to minimize neglected simulation time and cast the approach more broadly among a growing field of multi-fidelity methods.
Porcu, Roberto↗
Toward Transparent Data Management in Multi-layer Storage Hierarchy for HPC Systems
Upcoming exascale high performance computing (HPC) systems are expected to comprise multi-tier storage hierarchy, and thus will necessitate innovative storage and I/O mechanisms. Traditional disk and block-based interfaces and file systems face severe challenges in utilizing capabilities of storage hierarchies due to the lack of hierarchy support and semantic interfaces. Object-based and semantically-rich data abstractions for scientific data management on large scale systems offer a sustainable solution to these challenges. Such data abstractions can also simplify users involvement in data movement. Here, we take the first steps of realizing such an object abstraction and explore storage mechanisms for these objects to enhance I/O performance, especially for scientific applications. We explore how an object-based interface can facilitate next generation scalable computing systems by presenting the mapping of data I/O from two real world HPC scientific use cases: a plasma physics simulation code (VPIC) and a cosmology simulation code (HACC). Our storage model stores data objects in different physical organizations to support data movement across layers of memory/storage hierarchy. Our implementation sclaes well to 16K parallel processes, and compared to the state of the art, such as MPI-IO and HDF5, our object-based data abstractions and data placement strategy in multi-level storage hierarchy achieves up to 7 X I/O performance improvement for scientific data.
97 MATHEMATICS AND COMPUTING↗
Segmented Energy Routing for a Modular AC/DC Hybrid System
This article presents a modular ac/dc system with both distributed and centralized power ports for energy router (ER) applications. In each module of the described system, photovoltaic (PV) power generation units, battery-type energy storage (ES) units, and critical loads are connected to the cascaded H-bridge (CHB)-organized medium-voltage (MV) dc links, with fully distributed low-voltage (LV) dc power ports. Copies of modules share the centralized load bus and interact with an MV ac grid in parallel. Hybrid power port (HPO) assigns flexibility to the system but makes energy routings a necessity for stable operation. In this article, a segmented energy management strategy for the HPO-ER is proposed. In terms of the grid-side power transferring, the system ratings are intentionally designed to match significant power imbalance. Focally, a segmented energy management strategy is proposed to realize fully autonomous energy routing involving MV ac grid, distributed PV generations, distributed battery storages, and LV load. The system is proven to be stable using the derived impedance-based model considering the interaction between power blocks. The feasibility of the topology and control strategy is also verified through software simulation, laboratorial hardware prototype experiment, and hardware-in-the-loop (HIL) emulation.
24 POWER TRANSMISSION AND DISTRIBUTION↗
Fan-In Communications On A Cray Gemini Interconnect
Using the Cray Gemini interconnect as our platform, we present a study of an important class of communication operations the fan-in communication pattern. By its nature, fan-in communications form hot spots that present significant challenges for any interconnect fabric and communication software stack. Yet despite the inherent challenges, these communication patterns are common in both applications (which often perform reductions and other collective operations that include fan-in communication such as barriers) and system software (where they assume an important role within parallel file systems and other components requiring high-bandwidth or low-latency I/O). Our study determines the effectiveness of differing clientserver fan-in strategies. We describe fan-in performance in terms of aggregate bandwidth in the presence of varying degrees of congestion, as well as several other key attributes. Comparison numbers are presented for the Cray Aries interconnect. Finally, we provide recommended communication strategies based on our findings.
Jones, Terry R↗