SEARCH · Search NASA
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.
Privateer is a general-purpose data store that optimizes the tradeoff between storage space utilization and I/O performance. Privateer uses memory-mapped I/O with private mapping and an optimized writeback mechanism to maximize write parallelism and eliminate redundant writes; it also uses contentaddressable storage to optimize storage space via de-duplication.
This book constitutes the proceedings of the 4th OpenSHMEM Workshop, held in Annapolis, MD, USA, in August 2017.The 11 full papers presented in this book were carefully reviewed and selected from 14 submissions. The papers discuss a variety of ideas for extending the OpenSHMEM specification and making it efficient for current and next generation systems. This includes new research for communication contexts in OpenSHMEM, different optimizations for OpenSHMEM on shared memory machines, exploring the implementation of OpenSHMEM and its memory model on Intel’s KNL architecture, and implementing new applications and benchmarks with OpenSHMEM.
Not Available
We present algorithmic improvements to the loading operations of certain reduced data ensembles produced from neutron scattering experiments at Oak Ridge National Laboratory (ORNL) facilities. Ensembles from multiple measurements are required to cover a wide range of the phase space of a sample material of interest. They are stored using the standard NeXus schema on individual HDF5 files. This makes it a scalability challenge, as the number of experiments stored increases in a single ensemble file. The present work follows up on our previous efforts on data management algorithms, to address identified input output (I/O) bottlenecks in Mantid, an open-source data analysis framework used across several neutron science facilities around the world. We reuse an in-memory binary-tree metadata index that resembles data access patterns, to provide a scalable search and extraction mechanism. In addition, several memory operations are refactored and optimized for the current common use cases, ranging most frequently from 10 to 180, and up to 360 separate measurement configurations. Results from this work show consistent speed ups in wall-clock time on the Mantid LoadMD routine, ranging from 19% to 23% on average, on ORNL production computing systems. The latter depends on the complexity of the targeted instrument-specific data and the system I/O and compute variability for the shared computational resources available to users of ORNL’s Spallation Neutron Source (SNS) and the High Flux Isotope Reactor (HFIR) instruments. Nevertheless, we continue to highlight the need for more research to address reduction challenges as experimental data volumes, user time and processing costs increase.
Purpose: A major goal of an effective radiation treatment plan is to deliver the maximum dose to the tumor while minimizing radiation exposure to the surrounding normal structures. For example, due to the radiation exposure to neighboring critical structures during prostate cancer treatment, a significant increase in cancer risk was observed for the bladder (77%) and the rectum (105%) over the following decade. Consequently, an effective treatment plan necessitates limiting the exposure to such organs which can best be achieved by physically displacing the organ at-risk. The goal of this study is to present a prototype for an organ re-positioner device designed and fabricated to physically move the rectum away from the path of radiation beam during external beam and brachytherapy treatments. This device affords patient comfort and provides a fully controlled motion to safely relocate the rectum during treatment. Methods: The NiTi shape memory alloy was designed and optimized for manufacturing a rectal re-positioner device through cooling and heating the core alloy for its shaping. This has been achieved through a prototyped custom designed electronic circuit in order to induce the reversible austenitic transformation and was tested rigorously to ensure the integrity of the actuated motion in displacement of the target anatomy. Results: The desirable NiTi shape-setting was configured for easy insertion and based on anatomical constraint. When the final prototype was evaluated, accuracy and precision of the maximum displacement and temperature changes revealed that the device could safely be used within the target anatomy. Conclusion: The organ re-positioner device is a promising tool that can be implemented in clinical setting. It provides a controlled and safe displacement of the delicate organ(s) at risk. The location of the organ being treated could also be identified using conventional onboard imaging devices or MV imaging available on-board most modern clinical accelerators.
With each CMOS technology generation, leakage energy consumption has been dramatically increasing and hence, managing leakage power consumption of large last-level caches (LLCs) has become a critical issue in modern processor design. In this paper, we present EnCache, a novel software-based technique which uses dynamic profiling-based cache reconfiguration for saving cache leakage energy. EnCache uses a simple hardware component called profiling cache, which dynamically predicts energy efficiency of an application for 32 possible cache configurations. Using these estimates, system software reconfigures the cache to the most energy efficient configuration. EnCache uses dynamic cache reconfiguration and hence, it does not require offline profiling or tuning the parameter for each application. Furthermore, EnCache optimizes directly for the overall memory subsystem (LLC and main memory) energy efficiency instead of the LLC energy efficiency alone. The experiments performed with an x86-64 simulator and workloads from SPEC2006 suite confirm that EnCache provides larger energy saving than a conventional energy saving scheme. For single core and dual-core system configurations, the average savings in memory subsystem energy over a shared baseline configuration are 30.0% and 27.3%, respectively.
This research project has the objective to extend the range of application, improve the efficiency and conduct simulations with the Fast Lubrication Dynamics (FLD) algorithm for concentrated particle suspensions in a Newtonian fluid solvent. The research involves a combination of mathematical development, new computational algorithms, and application to processing flows of relevance in materials processing. The mathematical developments clarify the underlying theory, facilitate verification against classic monographs in the field and provide the framework for a novel parallel implementation optimized for an OpenMP shared memory environment. The project considered application to consolidation flows of major interest in high throughput materials processing and identified hitherto unforeseen challenges in the use of FLD in these applications. Extensions to the algorithm have been developed to improve its accuracy in these applications.
The field of bioinformatics and computational biology is currently experiencing a data revolution. The exciting prospect of making fundamental biological discoveries is fueling the rapid development and deployment of numerous cost-effective, high-throughput next-generation sequencing technologies. The result is that the DNA and protein sequence repositories are being bombarded with new sequence information. Databases are continuing to report a Moore’s law-like growth trajectory in their database sizes, roughly doubling every 18 months. In what seems to be a paradigm-shift, individual projects are now capable of generating billions of raw sequence data that need to be analyzed in the presence of already annotated sequence information. While it is clear that data-driven methods, such as sequencing homology detection, are becoming the mainstay in the field of computational life sciences, the algorithmic advancements essential for implementing complex data analytics at scale have mostly lagged behind. Sequence homology detection is central to a number of bioinformatics applications including genome sequencing and protein family characterization. Given millions of sequences, the goal is to identify all pairs of sequences that are highly similar (or “homologous”) on the basis of alignment criteria. While there are optimal alignment algorithms to compute pairwise homology, their deployment for large-scale is currently not feasible; instead, heuristic methods are used at the expense of quality. In this dissertation, we present the design and evaluation of a parallel implementation for conducting optimal homology detection on distributed memory supercomputers. Our approach uses a combination of techniques from asynchronous load balancing (viz. work stealing, dynamic task counters), data replication, and exact-matching filters to achieve homology detection at scale. Results for a collection of 2.56M sequences show parallel efficiencies of ~75-100% on up to 8K cores, representing a time-to-solution of 33 seconds. We extend this work with a detailed analysis of single-node sequence alignment performance using the latest CPU vector instruction set extensions. Preliminary results reveal that current sequence alignment algorithms are unable to fully utilize widening vector registers.
Deep Neural Networks (DNNs) have transformed the field of artificial intelligence and achieved extraordinary success in many areas. The training of DNNs is commonly compute and memory-intensive, which has resulted in several optimizations in the training phase. Among them, reduced precision is a typical and widely used technique to accelerate DNN training and reduce memory requirements. However, applying a widely adopted reduced precision format such as Float16 to all involved operations in DNN training is not optimal as the use of Float16 in some operations can hurt model accuracy. Meanwhile, additional optimizations including loss scaling and autocast techniques can mitigate the accuracy loss but lead to inherent overhead and inadequate use of reduced precision. In this work, we leverage another reduced precision format, BFloat16, and introduce a throughput-oriented and accuracy-aware approach to maximize the performance potential of DNN training. Since the high throughput provided by BFloat16 format is accompanied by low precision of the floating-point representation, this approach achieves high throughput by using BFloat16 on all DNN operations and avoids the accuracy loss through a customized accuracy-aware normalization. Results show that our approach outperforms the state-of-the-art mixed-precision training by 1.21x on an NVIDIA A100 GPU.
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.
The Eikonal equation is used to calculate wave propagation and distance fields, and due to its complexity requires numerical treatment for its solution. In this work, we present a second-order distributed memory parallel fast sweeping method. The second-order solution switches on a two-point stencil when two upwind points are available, and reverts to first-order otherwise. In all examples, the second-order method improves the solution over the first-order, allowing for significant savings in memory while achieving the same accuracy. Parallelization over distributed memory saw good weak scaling with optimal convergence. The computational time for second-order was approximately 2.5 times slower than first-order, where the largest amount of mesh points ran on 144 cores (512 GB) was ≈20 billion. The savings in memory from the second-order method combined with the distributed memory algorithm result in the ability to solve problems much larger than are possible with the serial first-order method.
Memory scalability is an enduring problem and bottleneck that plagues many parallel codes. Parallel codes designed for High Performance Systems are typically designed over the span of several, and in some instances 10+, years. As a result, optimization practices which were appropriate for earlier systems may no longer be valid and thus require careful optimization consideration. Specifically, parallel codes whose memory footprint is a function of their scalability must be carefully considered for future exa-scale systems. In this paper we present a methodology and tool to study the memory scalability of parallel codes. Using our methodology we evaluate an application s memory footprint as a function of scalability, which we coined memory efficiency, and describe our results. In particular, using our in-house tools we can pinpoint the specific application components which contribute to the application s overall memory foot-print (application data- structures, libraries, etc.).
An efficient parallel tensor transpose algorithm is suggested for shared-memory computing units, namely, multicore CPU, Intel Xeon Phi, and NVidia GPU. The algorithm operates on dense tensors (multidimensional arrays) and is based on the optimization of cache utilization on x86 CPU and the use of shared memory on NVidia GPU. From the applied side, the ultimate goal is to minimize the overhead encountered in the transformation of tensor contractions into matrix multiplications in computer implementations of advanced methods of quantum many-body theory (e.g., in electronic structure theory and nuclear physics). A particular accent is made on higher-dimensional tensors that typically appear in the so-called multireference correlated methods of electronic structure theory. Depending on tensor dimensionality, the presented optimized algorithms can achieve an order of magnitude speedup on x86 CPUs and 2-3 times speedup on NVidia Tesla K20X GPU with respect to the na ve scattering algorithm (no memory access optimization). Furthermore, the tensor transpose routines developed in this work have been incorporated into a general-purpose tensor algebra library (TAL-SH).
Many popular vetting tools for Android applications use static code analysis techniques. In particular, Interprocedural Data-Flow Graph (IDFG) construction is the computation at the core of Android static data-flow analysis and consumes most of the analysis time. Many analysis tools use a worklist algorithm, an iterative fixed-point approach, to construct the IDFG. In this paper, we observe that a straightforward GPU parallelization of the worklist algorithm leads to significant underutilization of the GPU resources. We identify four performance bottlenecks, namely, frequent dynamic memory allocations, high branch divergence, workload imbalance, and irregular memory access patterns. Accordingly, we propose GDroid, a GPU-based worklist algorithm implementation with multiple fine-grained optimizations tailored to common characteristics of Android applications. The optimizations considered are: matrix-based data structure, memory access-based node grouping, and worklist merging. Our experimental evaluation, performed on 1000 Android applications, shows that the proposed optimizations are beneficial to performance, and GDroid can achieve up to 128X speedups against a plain GPU implementation.
Physics-Informed Neural Networks (PINNs) have revolutionized the computation of PDE solutions by integrating partial differential equations (PDEs) into the neural network’s training process as soft constraints, becoming an important component of the scientific machine learning (SciML) ecosystem. More recently, physics-informed Kolmogorv-Arnold networks (PIKANs) have also shown to be effective and comparable in accuracy with PINNs. In their current implementation, both PINNs and PIKANs are mainly optimized using first-order methods like Adam, as well as quasi-Newton methods such as BFGS and its low-memory variant, L-BFGS. However, these optimizers often struggle with highly nonlinear and non-convex loss landscapes, leading to challenges such as slow convergence, local minima entrapment, and (non)degenerate saddle points. In this study, we investigate the performance of Self- Scaled BFGS (SSBFGS), Self-Scaled Broyden (SSBroyden) methods and other advanced quasi-Newton schemes, including BFGS and L-BFGS with different line search strategies. These methods dynamically rescale updates based on historical gradient information, thus enhancing training efficiency and accuracy. We systematically compare these optimizers – using both PINNs and PIKANs – on key challenging PDEs, including the Burgers, Allen-Cahn, Kuramoto-Sivashinsky, Ginzburg-Landau, and Stokes equations. Additionally, we evaluate the performance of SSBFGS and SSBroyden for Deep Operator Network (DeepONet) architectures, demonstrating their effectiveness for data-driven operator learning. Our findings provide state-of-the-art results with orders-of-magnitude accuracy improvements without the use of adaptive weights or any other enhancements typically employed in PINNs. More broadly, our work reveal insights into the effectiveness of quasi-Newton optimization strategies in significantly improving the convergence and accurate generalization of PINNs and PIKANs.
This paper proposes an efficient data memory management approach for the Intelligent RuntIme System (IRIS) heterogeneous computing framework along with new data transfer policies. IRIS provides a task-based programming model for extreme heterogeneous computing (e.g., CPU, GPU, DSP, FPGA) with support for today's most important programming languages (e.g., OpenMP, OpenCL, CUDA, HIP, OpenACC). However, the IRIS framework either forces the programmer to introduce data transfer commands for each task or relies on suboptimal memory management for automatic and transparent data transfers. The work described here extends IRIS with novel heterogeneous memory handling and introduces novel data transfer policies by employing the Distributed data MEMory handler (DMEM) for efficient and optimal movement of data among the various computing resources. The proposed approach achieves performance gains of up to 7× for tiled LU factorization and tiled DGEMM (i.e., matrix multiplication) benchmarks. Moreover, this approach also reduces data transfers by up to 71% when compared to previous IRIS heterogeneous memory management handlers. This work compares the performance results of the IRIS framework's novel DMEM with the StarPU runtime and MAGMA math library for GPUs. Experiments show a performance gain of up to 1.95× over StarPU and 2.1× over MAGMA.
The metal nanocrystals (NCs) embedded-NiN-based resistive random access memory cells are demonstrated using several metal NCs (i.e., Pt, Ni, and Ti) with different physical parameters in order to investigate the metal NC's dependence on resistive switching (RS) characteristics. First, depending on the electronegativity of metal, the size of metal NCs is determined and this affects the operating current of memory cells. If metal NCs with high electronegativity are incorporated, the size of the NCs is reduced; hence, the operating current is reduced owing to the reduced density of the electric field around the metal NCs. Second, the potential wells are formed by the difference of work function between the metal NCs and active layer, and the barrier height of the potential wells affects the level of operating voltage as well as the conduction mechanism of metal NCs embedded memory cells. Therefore, by understanding these correlations between the active layer and embedded metal NCs, we can optimize the RS properties of metal NCs embedded memory cells as well as predict their conduction mechanisms.
Here, this work extends our previous research entitled "MemXCT: Memory-centric X-ray CT Reconstruction with Massive Parallelization" that was originally published at SC19 conference (Hidayetoglu et al., 2019) with reproducibility of the computational imaging performance. X-ray computed tomography (XCT) is regularly used at synchrotron light sources to study the internal morphology of materials at high resolution. However, experimental constraints, such as radiation sensitivity, can result in noisy or undersampled measurements. Further, depending on the resolution, sample size and data acquisition rates, the resulting noisy dataset can be in the order of terabytes. Advanced iterative reconstruction techniques can produce high-quality images from noisy measurements, but their computational requirements have made their use an exception rather than the rule. We propose a novel memory-centric approach that avoids redundant computations at the expense of additional memory complexity. We develop a memory-centric iterative reconstruction system, MemXCT, that uses an optimized SpMV implementation with two-level pseudo-Hilbert ordering and multi-stage input buffering. We evaluate MemXCT on various supercomputer architectures involving KNL and GPU. MemXCT can reconstruct a large (11Kx11K) mouse brain tomogram in 10 seconds using 4096 KNL nodes (256K cores). The results presented in our original article at the SC19 were based on large-scale supercomputing resources. The MemXCT application was selected for the Student Cluster Competition (SCC) Reproducibility Challenge and evaluated on a variety of cloud computing resources by universities around the world in the SC20 conference. We summarize the results of the top-ranked SCC Reproducibility Challenge teams and identify the most pertinent measures for ensuring the reproducibility of our experiments in this article.