Search NASA⌕ Search

SEARCH · Search NASA

Results for “Memory systems”

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 1,657 records · Page 92

Production Application Performance Data Streaming for System Monitoring

In this article, we present an approach to streaming collection of application performance data. Practical application performance tuning and troubleshooting in production high-performance computing (HPC) environments requires an understanding of how applications interact with the platform, including (but not limited to) parallel programming libraries such as Message Passing Interface (MPI). Several profiling and tracing tools exist that collect heavy runtime data traces either in memory (released only at application exit) or on a file system (imposing an I/O load that may interfere with the performance being measured). Although these approaches are beneficial in development stages and post-run analysis, a systemwide and low-overhead method is required to monitor deployed applications continuously. This method must be able to collect information at both the application and system levels to yield a complete performance picture. In our approach, an application profiler collects application event counters. A sampler uses an efficient inter-process communication method to periodically extract the application counters and stream them into an infrastructure for performance data collection. We implement a tool-set based on our approach and integrate it with the Lightweight Distributed Metric Service (LDMS) system, a monitoring system used on large-scale computational platforms. LDMS provides the infrastructure to create and gather streams of performance data in a low overhead manner. We demonstrate our approach using applications implemented with MPI, as it is one of the most common standards for the development of large-scale scientific applications. We utilize our tool-set to study the impact of our approach on an open source HPC application, Nalu. Our tool-set enables us to efficiently identify patterns in the behavior of the application without source-level knowledge. We leverage LDMS to collect system-level performance data and explore the correlation between the system and application events. Finally, we demonstrate how our tool-set can help detect anomalies with a low latency. We run tests on two different architectures: a system enabled with Intel Xeon Phi and another system equipped with Intel Xeon processor. Our overhead study shows our method imposes at most 0.5% CPU usage overhead on the application in realistic deployment scenarios.

59 BASIC BIOLOGICAL SCIENCES↗

Multi-core processing and scheduling performance in CMS

Commodity hardware is going many-core. We might soon not be able to satisfy the job memory needs per core in the current single-core processing model in High Energy Physics. In addition, an ever increasing number of independent and incoherent jobs running on the same physical hardware not sharing resources might significantly affect processing performance. It will be essential to effectively utilize the multi-core architecture. CMS has incorporated support for multi-core processing in the event processing framework and the workload management system. Multi-core processing jobs share common data in memory, such us the code libraries, detector geometry and conditions data, resulting in a much lower memory usage than standard single-core independent jobs. Exploiting this new processing model requires a new model in computing resource allocation, departing from the standard single-core allocation for a job. The experiment job management system needs to have control over a larger quantum of resource since multi-core aware jobs require the scheduling of multiples cores simultaneously. CMS is exploring the approach of using whole nodes as unit in the workload management system where all cores of a node are allocated to a multi-core job. Whole-node scheduling allows for optimization of the data/workflow management (e.g. I/O caching, local merging) but efficient utilization of all scheduled cores is challenging. Dedicated whole-node queues have been setup at all Tier-1 centers for exploring multi-core processing workflows in CMS. We present the evaluation of the performance scheduling and executing multi-core workflows in whole-node queues compared to the standard single-core processing workflows.

Hernandez, J. M.↗

On Building Predictive Digital Twin Incorporating Wave Predicting Capabilities: Case Study on UMaine Experimental Campaign - FOCAL

The response of floating wind turbines (FWT) are susceptible to stochastic wave variations. For the optimal operation of FWT, a comprehensive understanding of the phaseresolved wave dynamics and the consequential system response is crucial for real-time monitoring and control. A multi-variate, multi-step, long short term memory (MLSTM), a type of recurrent neural network (RNN) is used to capture complex system dynamics for real-time application. Results indicate that the integration of a wave prediction-reconstruction (WRP) model substantially enhances prediction accuracy by 50% on average relative to the baseline model. The improvement is consistent across various wave extremity and prediction horizons, thereby significantly broadening the scope for timely and precise predictive capabilities.

17 WIND ENERGY↗

SU-E-J-60: Efficient Monte Carlo Dose Calculation On CPU-GPU Heterogeneous Systems

Purpose: It is well-known that the performance of GPU-based Monte Carlo dose calculation implementations is bounded by memory bandwidth. One major cause of this bottleneck is the random memory writing patterns in dose deposition, which leads to several memory efficiency issues on GPU such as un-coalesced writing and atomic operations. We propose a new method to alleviate such issues on CPU-GPU heterogeneous systems, which achieves overall performance improvement for Monte Carlo dose calculation. Methods: Dose deposition is to accumulate dose into the voxels of a dose volume along the trajectories of radiation rays. Our idea is to partition this procedure into the following three steps, which are fine-tuned for CPU or GPU: (1) each GPU thread writes dose results with location information to a buffer on GPU memory, which achieves fully-coalesced and atomic-free memory transactions; (2) the dose results in the buffer are transferred to CPU memory; (3) the dose volume is constructed from the dose buffer on CPU. We organize the processing of all radiation rays into streams. Since the steps within a stream use different hardware resources (i.e., GPU, DMA, CPU), we can overlap the execution of these steps for different streams by pipelining. Results: We evaluated our method using a Monte Carlo Convolution Superposition (MCCS) program and tested our implementation for various clinical cases on a heterogeneous system containing an Intel i7 quad-core CPU and an NVIDIA TITAN GPU. Comparing with a straightforward MCCS implementation on the same system (using both CPU and GPU for radiation ray tracing), our method gained 2-5X speedup without losing dose calculation accuracy. Conclusion: The results show that our new method improves the effective memory bandwidth and overall performance for MCCS on the CPU-GPU systems. Our proposed method can also be applied to accelerate other Monte Carlo dose calculation approaches. This research was supported in part by NSF under Grants CCF-1217906, and also in part by a research contract from the Sandia National Laboratories.

60 APPLIED LIFE SCIENCES↗

Flexible Peripheral Component Interconnect Input/Output Card

The Flexible Peripheral Component Interconnect (PCI) Input/Output (I/O) Card is an innovative circuit board that provides functionality to interface between a variety of devices. It supports user-defined interrupts for interface synchronization, tracks system faults and failures, and includes checksum and parity evaluation of interface data. The card supports up to 16 channels of high-speed, half-duplex, low-voltage digital signaling (LVDS) serial data, and can interface combinations of serial and parallel devices. Placement of a processor within the field programmable gate array (FPGA) controls an embedded application with links to host memory over its PCI bus. The FPGA also provides protocol stacking and quick digital signal processor (DSP) functions to improve host performance. Hardware timers, counters, state machines, and other glue logic support interface communications. The Flexible PCI I/O Card provides an interface for a variety of dissimilar computer systems, featuring direct memory access functionality. The card has the following attributes: 8/16/32-bit, 33-MHz PCI r2.2 compliance, Configurable for universal 3.3V/5V interface slots, PCI interface based on PLX Technology's PCI9056 ASIC, General-use 512K 16 SDRAM memory, General-use 1M 16 Flash memory, FPGA with 3K to 56K logical cells with embedded 27K to 198K bits RAM, I/O interface: 32-channel LVDS differential transceivers configured in eight, 4-bit banks; signaling rates to 200 MHz per channel, Common SCSI-3, 68-pin interface connector.

Bigelow, Kirk K.↗

Scaling Deep Learning on GPU and Knights Landing clusters

The speed of deep neural networks training has become a big bottleneck of deep learning research and development. For example, training GoogleNet by ImageNet dataset on one Nvidia K20 GPU needs 21 days. To speed up the training process, the current deep learning systems heavily rely on the hardware accelerators. However, these accelerators have limited on-chip memory compared with CPUs. To handle large datasets, they need to fetch data from either CPU memory or remote processors. We use both self-hosted Intel Knights Landing (KNL) clusters and multi-GPU clusters as our target platforms. From an algorithm aspect, current distributed machine learning systems are mainly designed for cloud systems. These methods are asynchronous because of the slow network and high fault-tolerance requirement on cloud systems. We focus on Elastic Averaging SGD (EASGD) to design algorithms for HPC clusters. Original EASGD used round-robin method for communication and updating. The communication is ordered by the machine rank ID, which is inefficient on HPC clusters. First, we redesign four efficient algorithms for HPC systems to improve EASGD's poor scaling on clusters. Async EASGD, Async MEASGD, and Hogwild EASGD are faster \textcolor{black}{than} their existing counterparts (Async SGD, Async MSGD, and Hogwild SGD, resp.) in all the comparisons. Finally, we design Sync EASGD, which ties for the best performance among all the methods while being deterministic. In addition to the algorithmic improvements, we use some system-algorithm codesign techniques to scale up the algorithms. By reducing the percentage of communication from 87% to 14%, our Sync EASGD achieves 5.3x speedup over original EASGD on the same platform. We get 91.5% weak scaling efficiency on 4253 KNL cores, which is higher than the state-of-the-art implementation.

60 APPLIED LIFE SCIENCES↗

Forces and stress in second order Møller-Plesset perturbation theory for condensed phase systems within the resolution-of-identity Gaussian and plane waves approach

The forces acting on the atoms as well as the stress tensor are crucial ingredients for calculating the structural and dynamical properties of systems in the condensed phase. Here, these derivatives of the total energy are evaluated for the second-order Møller-Plesset perturbation energy (MP2) in the framework of the resolution of identity Gaussian and plane waves method, in a way that is fully consistent with how the total energy is computed. This consistency is non-trivial, given the different ways employed to compute Coulomb, exchange, and canonical four center integrals, and allows, for example, for energy conserving dynamics in various ensembles. Based on this formalism, a massively parallel algorithm has been developed for finite and extended system. The designed parallel algorithm displays, with respect to the system size, cubic, quartic, and quintic requirements, respectively, for the memory, communication, and computation. All these requirements are reduced with an increasing number of processes, and the measured performance shows excellent parallel scalability and efficiency up to thousands of nodes. Additionally, the computationally more demanding quintic scaling steps can be accelerated by employing graphics processing units (GPU’s) showing, for large systems, a gain of almost a factor two compared to the standard central processing unit-only case. In this way, the evaluation of the derivatives of the RI-MP2 energy can be performed within a few minutes for systems containing hundreds of atoms and thousands of basis functions. With good time to solution, the implementation thus opens the possibility to perform molecular dynamics (MD) simulations in various ensembles (microcanonical ensemble and isobaric-isothermal ensemble) at the MP2 level of theory. Geometry optimization, full cell relaxation, and energy conserving MD simulations have been performed for a variety of molecular crystals including NH{sub 3}, CO{sub 2}, formic acid, and benzene.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

BlackjackBench: Portable Hardware Characterization

DARPA's AACE project aimed to develop Architecture Aware Compiler Environments that automatically characterize the hardware and optimize the application codes accordingly. We present the BlackjackBench suite, a collection of portable micro-benchmarks that automate system characterization, plus statistical analysis techniques for interpreting the results. BlackjackBench discovers the effective sizes and speeds of the hardware environment rather than the often unattainable peak values. We aim at hardware features that can be observed by running executables generated by existing compilers from standard C codes. We characterize the memory hierarchy, including cache sharing and NUMA characteristics of the system, properties of the processing cores affecting execution speed, and the length of the OS scheduler time slot. We show how these features of modern multicores can be discovered programmatically. We also show how the features could interfere with each other resulting in incorrect interpretation of the results, and how established classification and statistical analysis techniques reduce experimental noise and aid automatic interpretation of results.

99 GENERAL AND MISCELLANEOUS↗

Search for the Heisenberg spin glass on rewired square lattices with antiferromagnetic interaction

Spin glass (SG) is a typical magnetic system with frozen random spin orientation at low temperatures. The system exhibits rich physical properties, such as infinite number of ground states, memory effect, and aging phenomena. There are two main ingredients considered to be pivotal for the existence of SG behavior, namely, frustration and randomness. For the canonical SG system, frustration is led by the presence of competing interaction between ferromagnetic (FM) and antiferromagnetic (AF) couplings. Previously, Bartolozzi et al. [Phys. Rev. B73, 224419 (2006)], reported the SG properties of the AF Ising spins on scale free network (SFN). It is a new type of SG, different from the canonical one which requires the presence of both FM and AF couplings. In this new system, frustration is purely caused by the topological factor and its randomness is related to the irregular connectvity. Recently, Surungan et. al. [Journal of Physics: Conference Series, 640, 012001 (2015)] reported SG bahavior of AF Heisenberg model on SFN. We further investigate this type of system by studying an AF Heisenberg model on rewired square lattices. We used Replica Exchange algorithm of Monte Carlo Method and calculated the SG order parameter to search for the existence of SG phase.

36 MATERIALS SCIENCE↗

Kinetics of dense matter: Correlations and memory

A link between correlation contributions to kinetic equations for dense quantum systems and the energy conservation is considered. In order that the energy be conserved by an approximate collision integral, the one-particle density matrix and the mean interaction energy are treated as independent state parameters. It is shown how the density operator method can be used to derive non-Markovian kinetic equations including the correlation effect associated with the energy conservation. A quantum generalization of the Enskog theory is discussed.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A comparison of select image-compression algorithms for an electronic still camera

This effort is a study of image-compression algorithms for an electronic still camera. An electronic still camera can record and transmit high-quality images without the use of film, because images are stored digitally in computer memory. However, high-resolution images contain an enormous amount of information, and will strain the camera's data-storage system. Image compression will allow more images to be stored in the camera's memory. For the electronic still camera, a compression algorithm that produces a reconstructed image of high fidelity is most important. Efficiency of the algorithm is the second priority. High fidelity and efficiency are more important than a high compression ratio. Several algorithms were chosen for this study and judged on fidelity, efficiency and compression ratio. The transform method appears to be the best choice. At present, the method is compressing images to a ratio of 5.3:1 and producing high-fidelity reconstructed images.

Nerheim, Rosalee↗

Numerical integration of the extended variable generalized Langevin equation with a positive Prony representable memory kernel

Generalized Langevin dynamics (GLD) arise in the modeling of a number of systems, ranging from structured fluids that exhibit a viscoelastic mechanical response, to biological systems, and other media that exhibit anomalous diffusive phenomena. Molecular dynamics (MD) simulations that include GLD in conjunction with external and/or pairwise forces require the development of numerical integrators that are efficient, stable, and have known convergence properties. In this article, we derive a family of extended variable integrators for the Generalized Langevin equation with a positive Prony series memory kernel. Using stability and error analysis, we identify a superlative choice of parameters and implement the corresponding numerical algorithm in the LAMMPS MD software package. Salient features of the algorithm include exact conservation of the first and second moments of the equilibrium velocity distribution in some important cases, stable behavior in the limit of conventional Langevin dynamics, and the use of a convolution-free formalism that obviates the need for explicit storage of the time history of particle velocities. Capability is demonstrated with respect to accuracy in numerous canonical examples, stability in certain limits, and an exemplary application in which the effect of a harmonic confining potential is mapped onto a memory kernel.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

FFTs in external or hierarchical memory

A description is given of advanced techniques for computing an ordered FFT on a computer with external or hierarchical memory. These algorithms (1) require as few as two passes through the external data set, (2) use strictly unit stride, long vector transfers between main memory and external storage, (3) require only a modest amount of scratch space in main memory, and (4) are well suited for vector and parallel computation. Performance figures are included for implementations of some of these algorithms on Cray supercomputers. Of interest is the fact that a main memory version outperforms the current Cray library FFT routines on the Cray-2, the Cray X-MP, and the Cray Y-MP systems. Using all eight processors on the Cray Y-MP, this main memory routine runs at nearly 2 Gflops.

Bailey, David H.↗

Analysis and Benchmarking of feature reduction for classification under computational constraints

Abstract Machine learning is most often expensive in terms of computational and memory costs due to training with large volumes of data. Current computational limitations of many computing systems motivate us to investigate practical approaches, such as feature selection and reduction, to reduce the time and memory costs while not sacrificing the accuracy of classification algorithms. In this work, we carefully review, analyze, and identify the feature reduction methods that have low costs/overheads in terms of time and memory. Then, we evaluate the identified reduction methods in terms of their impact on the accuracy, precision, time, and memory costs of traditional classification algorithms. Specifically, we focus on the least resource intensive feature reduction methods that are available in Scikit-Learn library. Since our goal is to identify the best performing low-cost reduction methods, we do not consider complex expensive reduction algorithms in this study. In our evaluation, we find that at quadratic-scale feature reduction, the classification algorithms achieve the best trade-off among competitive performance metrics. Results show that the overall training times are reduced 61%, the model sizes are reduced 6×, and accuracy scores increase 25% compared to the baselines on average with quadratic scale reduction.

97 MATHEMATICS AND COMPUTING↗

Integral Kernel Methods for Nonlinear Parabolic-Elliptic Systems

Nonlinear parabolic-elliptic systems arise in many physical, biological, and chemical phenomena such as chemotaxis, ion transport, self-gravitating particles, and Brownian vortices. Existing methods struggle with the strong coupling and high nonlinearity and nonlocality of some of these systems, especially the ill-conditioned, convection-dominated problems. To overcome numerical difficulties, current approaches rely on initial guesses, preconditioning, or iterative techniques with no convergence guarantees. They might suffer from poor scalability, large memory usage, and difficulty to parallelize. Inspired by the connection of parabolic-elliptic systems to stochastic processes, we introduce a novel meshless, monolithic, and fully explicit method that naturally encapsulates the elliptic and parabolic operators into a single step which updates each node deterministically with global information. By being fully quadrature-based, it avoids solving systems of discretized equations and does not utilize initial guesses or preconditioning, while requiring little memory and being easy to parallelize. We first derive the method in an integral kernel formulation with quadratic complexity in the number of integration nodes and then leverage kernel-independent fast multipole methods (FMM) to present a scalable algorithm with linear complexity. We provide numerical examples for the Poisson-Nernst-Planck equations in one, two, and three dimensions, together with the derivation of the integral kernel for each case. Furthermore, the examples demonstrate the fast convergence and scalability of the FMM-accelerated algorithm, as well as its suitability for convection-dominated problems, making it competitive against traditional PDE solvers.

PDE systems↗

Rolex: Resilience-oriented language extensions for extreme-scale systems

Future exascale high-performance computing (HPC) systems will be constructed from VLSI devices that will be less reliable than those used today, and faults will become the norm, not the exception. This will pose significant problems for system designers and programmers, who for half-a-century have enjoyed an execution model that assumed correct behavior by the underlying computing system. The mean time to failure (MTTF) of the system scales inversely to the number of components in the system and therefore faults and resultant system level failures will increase, as systems scale in terms of the number of processor cores and memory modules used. However every error detected need not cause catastrophic failure. Many HPC applications are inherently fault resilient. Yet it is the application programmers who have this knowledge but lack mechanisms to convey it to the system. In this paper, we present new Resilience Oriented Language Extensions (Rolex) which facilitate the incorporation of fault resilience as an intrinsic property of the application code. We describe the syntax and semantics of the language extensions as well as the implementation of the supporting compiler infrastructure and runtime system. Furthermore, our experiments show that an approach that leverages the programmer's insight to reason about the context and significance of faults to the application outcome significantly improves the probability that an application runs to a successful conclusion.

97 MATHEMATICS AND COMPUTING↗

Improving scalability of electronic structure code for molecular simulations in the presence of environment

A scalable density functional electronic code with Gaussian basis set, called UTEP-NRLMOL, is developed to perform simulations of molecular systems in the presence of the environment with particular attention to the memory requirements. In the electronic structure calculations, the memory and computation time are proportional to the number of atoms. Memory requirements for density functional calculations scale as N*N, where N is the number of atoms. While the recent advances in HPC offer platforms with large numbers of cores, the limited amount of memory available on a given node and poor scalability of the electronic structure codes hinder their efficient usage of these platforms. We have introduced new scaling and parallelization paradigms using MPI-3 shared-memory functionality combined with usage of sparse algebra and storage of matrices in sparse format. This extends the range of applicability of the UTEP-NRLMOL code to large systems over 10,000 atoms, or using up to 67,000 basis functions, and making use of HPC architectures using over 6,000 processors utilizing all available cores. We have also interfaced code with effective fragment potential and polarizable continuum model libraries. The code was used in simulations of several applications which are published in reputed scientific journals.

74 ATOMIC AND MOLECULAR PHYSICS↗

A Monte Carlo neutron transport code for eigenvalue calculations on a dual-GPU system and CUDA environment

Monte Carlo (MC) method is able to accurately calculate eigenvalues in reactor analysis. Its lengthy computation time can be reduced by general-purpose computing on Graphics Processing Units (GPU), one of the latest parallel computing techniques under development. The method of porting a regular transport code to GPU is usually very straightforward due to the 'embarrassingly parallel' nature of MC code. However, the situation becomes different for eigenvalue calculation in that it will be performed on a generation-by-generation basis and the thread coordination should be explicitly taken care of. This paper presents our effort to develop such a GPU-based MC code in Compute Unified Device Architecture (CUDA) environment. The code is able to perform eigenvalue calculation under simple geometries on a multi-GPU system. The specifics of algorithm design, including thread organization and memory management were described in detail. The original CPU version of the code was tested on an Intel Xeon X5660 2.8 GHz CPU, and the adapted GPU version was tested on NVIDIA Tesla M2090 GPUs. Double-precision floating point format was used throughout the calculation. The result showed that a speedup of 7.0 and 33.3 were obtained for a bare spherical core and a binary slab system respectively. The speedup factor was further increased by a factor of {approx}2 on a dual GPU system. The upper limit of device-level parallelism was analyzed, and a possible method to enhance the thread-level parallelism was proposed. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗