Search NASA⌕ Search

SEARCH · Search NASA

Results for “memory efficiency”

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 91 records · Page 5

Metall: A persistent memory allocator for data-centric analytics

Data analytics applications transform raw input data into analytics-specific data structures before performing analytics. Unfortunately, such data ingestion steps are often more expensive than analytics. In addition, various types of NVRAM devices are already used in many HPC systems today. Such devices will be useful for storing and reusing data structures beyond a single process life cycle. We developed Metall, a persistent memory allocator built on top of the memory-mapped file mechanism. Metall enables applications to transparently allocate custom C++ data structures into various types of persistent memories. Metall incorporates a concise and high-performance memory management algorithm inspired by Supermalloc and the rich C++ interface developed by Boost.Interprocess library. On a dynamic graph construction workload, Metall achieved up to 11.7x and 48.3x performance improvements over Boost.Interprocess and memkind (PMEM kind), respectively. We also demonstrate Metall’s high adaptability by integrating Metall into a graph processing framework, GraphBLAS Template Library. Here this study’s outcomes indicate that Metall will be a strong tool for accelerating future large-scale data analytics by allowing applications to leverage persistent memory efficiently.

97 MATHEMATICS AND COMPUTING↗

Optimizing Performance of Combustion Chemistry Solvers on Intel's Many Integrated Core (MIC) Architectures

This work investigates novel algorithm designs and optimization techniques for restructuring chemistry integrators in zero and multidimensional combustion solvers, which can then be effectively used on the emerging generation of Intel's Many Integrated Core/Xeon Phi processors. These processors offer increased computing performance via large number of lightweight cores at relatively lower clock speeds compared to traditional processors (e.g. Intel Sandybridge/Ivybridge) used in current supercomputers. This style of processor can be productively used for chemistry integrators that form a costly part of computational combustion codes, in spite of their relatively lower clock speeds. Performance commensurate with traditional processors is achieved here through the combination of careful memory layout, exposing multiple levels of fine grain parallelism and through extensive use of vendor supported libraries (Cilk Plus and Math Kernel Libraries). Important optimization techniques for efficient memory usage and vectorization have been identified and quantified. These optimizations resulted in a factor of ~ 3 speed-up using Intel 2013 compiler and ~ 1.5 using Intel 2017 compiler for large chemical mechanisms compared to the unoptimized version on the Intel Xeon Phi. The strategies, especially with respect to memory usage and vectorization, should also be beneficial for general purpose computational fluid dynamics codes.

97 MATHEMATICS AND COMPUTING↗

Learning Constrained Parametric Differentiable Predictive Control Policies With Guarantees

We present differentiable predictive control (DPC), a method for offline learning of constrained neural control policies for nonlinear dynamical systems with performance guarantees. We show that the sensitivities of the parametric optimal control problem can be used to obtain direct policy gradients. Specifically, we employ automatic differentiation (AD) to efficiently compute the sensitivities of the model predictive control (MPC) objective function and constraints penalties. To guarantee safety upon deployment, we derive probabilistic guarantees on closed-loop stability and constraint satisfaction based on indicator functions and Hoeffding’s inequality. We empirically demonstrate that the proposed method can learn neural control policies for various parametric optimal control tasks. In particular, we show that the proposed DPC method can stabilize systems with unstable dynamics, track time-varying references, and satisfy nonlinear state and input constraints. Our DPC method has practical time savings compared to alternative approaches for fast and memory-efficient controller design. Specifically, DPC does not depend on a supervisory controller as opposed to approximate MPC based on imitation learning. We demonstrate that, without losing performance, DPC is scalable with greatly reduced demands on memory and computation compared to implicit and explicit MPC while being more sample efficient than model-free reinforcement learning (RL) algorithms.

97 MATHEMATICS AND COMPUTING↗

A simple implementation of the Viterbi algorithm on the Motorola DSP56001

As systems designers design communication systems with digital rather than analog components to reduce noise and increase channel capacity, they must have the ability to perform traditional communication algorithms digitally. The use of trellis coded modulation as well as the extensive use of convolutional encoding for error detection and correction requires an efficient digital implementation of the Viterbi Algorithm for real time demodulation and decoding. Digital signal processors are now fast enough to implement Viterbi decoding in conjunction with the normal receiver/transmitter functions for lower speed channels on a single chip as well as performing fast decoding for higher speed channels, if the algorithm is implemented efficiently. The purpose of this paper is to identify a good way to implement the Viterbi Algorithm (VA) on the Motorola DSP56001, balancing performance considerations with speed and memory efficiency.

Messer, Dion D.↗

Reduced-order modeling for efficient cross section library development in high-temperature gas reactor pebble-bed depletion analysis

Accurate modeling of running-in and equilibrium conditions in pebble-bed reactors (PBRs) requires precise microscopic multigroup neutron cross sections. In Griffin, deterministic neutronics calculations rely on multivariate interpolation over large cross section libraries, resulting in significant memory usage and performance bottlenecks. This work, together with a companion paper on Griffin integration, explores reduced-order models (ROMs) to replace interpolation with lightweight surrogates. Several ROM techniques are benchmarked, with deep neural networks (DNNs) demonstrating superior memory efficiency, scalability, and predictive accuracy. A total of 295 DNNs were trained to build a comprehensive isotope library, integrated into Griffin through a custom LibTorch interface for depletion analysis. Initial results demonstrate that DNN-based ROMs drastically reduce memory demands while preserving accuracy, enabling finer tabulations and additional state variables without overhead. In conclusion, the framework also supports online cross section generation and real-time DNN updates through transfer learning, improving fidelity by capturing self-shielding and evolving nuclide compositions during burnup.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗

A Parallel Cartesian Approach for External Aerodynamics of Vehicles with Complex Geometry

This workshop paper presents the current status in the development of a new approach for the solution of the Euler equations on Cartesian meshes with embedded boundaries in three dimensions on distributed and shared memory architectures. The approach uses adaptively refined Cartesian hexahedra to fill the computational domain. Where these cells intersect the geometry, they are cut by the boundary into arbitrarily shaped polyhedra which receive special treatment by the solver. The presentation documents a newly developed multilevel upwind solver based on a flexible domain-decomposition strategy. One novel aspect of the work is its use of space-filling curves (SFC) for memory efficient on-the-fly parallelization, dynamic re-partitioning and automatic coarse mesh generation. Within each subdomain the approach employs a variety reordering techniques so that relevant data are on the same page in memory permitting high-performance on cache-based processors. Details of the on-the-fly SFC based partitioning are presented as are construction rules for the automatic coarse mesh generation. After describing the approach, the paper uses model problems and 3- D configurations to both verify and validate the solver. The model problems demonstrate that second-order accuracy is maintained despite the presence of the irregular cut-cells in the mesh. In addition, it examines both parallel efficiency and convergence behavior. These investigations demonstrate a parallel speed-up in excess of 28 on 32 processors of an SGI Origin 2000 system and confirm that mesh partitioning has no effect on convergence behavior.

Aftosmis, M. J.↗

Adapting CLUTCH methodology to multigroup TSUNAMI-3D for eigenvalue sensitivity calculations

The sensitivity of the eigenvalue to uncertainties in nuclear data and its evaluation are important for nuclear criticality safety. TSUNAMI-3D sequences within the SCALE code system offer several options to the user community for calculating eigenvalue sensitivity coefficients with multigroup (MG) and continuous energy (CE) 3D transport capabilities. TSUNAMI-3D sequences implement the adjoint-based perturbation theory with MG KENO code, the Contributon Linked eigenvalue sensitivity/Uncertainty estimation via Track length importance CHaracterization (CLUTCH) method with CE KENO code, and the Iterated Fission Probability (IFP) method with CE KENO and Shift codes. Each method has benefits and limitations depending on the problem that is run. The work presented here aims to adapt the CLUTCH method, which enables the Contributon method's mesh-free, memory-efficient approach for calculating adjoint-weighted tallies for sensitivity calculations, to the MG TSUNAMI-3D sequence. This application would eliminate the explicit adjoint KENO calculation, as well as the memory-consuming mesh flux moment tallies required by the conventional MG TSUNAMI-3D. Smaller memory footprints in the CLUTCH methodology and relatively shorter runtimes in MG KENO transport can make MG TSUNAMI-3D a viable method for some complex problems. Moreover, this adaptation allows MG sensitivity calculations with Shift, ORNL's next-generation high-performance Monte Carlo transport code, which currently does not offer any sensitivity capabilities with MG particle transport simulations. Initial implementation of the new MG TSUNAMI-3D sequence and its preliminary results with a selected critical benchmark experiment in the Verified, Archived Library of Inputs and Data (VALID) are presented in this study.

KENO↗

An Evaluation of OpenSHMEM Interfaces for the Variable-Length Alltoallv() Collective Operation, In: OpenSHMEM and Related Technologies. Experiences, Implementations, and Technologies

Alltoallv() is a collective operation which allows all processes to exchange variable amounts of data with all other processes in the communication group. This means that Alltoallv() requires not only O(N2) communications, but typically also additional exchanges of the data lengths that will be transmitted in the eventual Alltoallv() call. This pre-exchange is used to calculate the proper offsets for the receiving buffers on the target processes. However, we propose two new candidate interfaces for Alltoallv() that would mitigate the need for the user to set up this extra exchange of information at the possible cost of memory efficiency. We explain the new interface variants and show how a single call can be used in place of the traditional Alltoall()/ Alltoallv() pair. We then discuss the performance tradeoffs for overall communication and memory costs, as well as both software and hardware-based optimizations and their applicability to the various proposed interfaces.

Computer Science↗

Linear-scaling time-dependent density-functional theory beyond the Tamm-Dancoff approximation: Obtaining efficiency and accuracy with in situ optimised local orbitals

We present a solution of the full time-dependent density-functional theory (TDDFT) eigenvalue equation in the linear response formalism exhibiting a linear-scaling computational complexity with system size, without relying on the simplifying Tamm-Dancoff approximation (TDA). The implementation relies on representing the occupied and unoccupied subspaces with two different sets of in situ optimised localised functions, yielding a very compact and efficient representation of the transition density matrix of the excitation with the accuracy associated with a systematic basis set. The TDDFT eigenvalue equation is solved using a preconditioned conjugate gradient algorithm that is very memory-efficient. The algorithm is validated on a small test molecule and a good agreement with results obtained from standard quantum chemistry packages is found, with the preconditioner yielding a significant improvement in convergence rates. The method developed in this work is then used to reproduce experimental results of the absorption spectrum of bacteriochlorophyll in an organic solvent, where it is demonstrated that the TDA fails to reproduce the main features of the low energy spectrum, while the full TDDFT equation yields results in good qualitative agreement with experimental data. Furthermore, the need for explicitly including parts of the solvent into the TDDFT calculations is highlighted, making the treatment of large system sizes necessary that are well within reach of the capabilities of the algorithm introduced here. Finally, the linear-scaling properties of the algorithm are demonstrated by computing the lowest excitation energy of bacteriochlorophyll in solution. The largest systems considered in this work are of the same order of magnitude as a variety of widely studied pigment-protein complexes, opening up the possibility of studying their properties without having to resort to any semiclassical approximations to parts of the protein environment.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

An Evaluation of OpenSHMEM Interfaces for the Variable-length Alltoallv() Collective Operation

Alltoallv() is a collective operation which allows all pro- cesses to exchange variable amounts of data with all other processes in the communication group. This means that Alltoallv() requires not only O(N2) communications, but typically also additional exchanges of the data lengths that will be transmitted in the eventual Alltoallv() call. This pre-exchange is used to calculate the proper offsets for the re- ceiving buffers on the target processes. However, we propose two new can- didate interfaces for Alltoallv() that would mitigate the need for the user to set up this extra exchange of information at the possible cost of memory efficiency. We explain the new interface variants and show how a single call can be used in place of the traditional Alltoall()/Alltoallv() pair. We then discuss the performance tradeoffs for overall communica- tion and memory costs, as well as both software and hardware-based optimizations and their applicability to the various proposed interfaces.

Lopez, Matthew Graham↗

Distributed Tomographic Reconstruction with Quantization

Conventional tomographic reconstruction typically depends on centralized servers for both data storage and computation, leading to concerns about memory limitations and data privacy. Distributed reconstruction algorithms mitigate these issues by partitioning data across multiple nodes, reducing server load and enhancing privacy. However, these algorithms often encounter challenges related to memory constraints and communication overhead between nodes. In this paper, we introduce a decentralized Alternating Directions Method of Multipliers (ADMM) with configurable quantization. By distributing local objectives across nodes, our approach is highly scalable and can efficiently reconstruct images while adapting to available resources. To overcome communication bottlenecks, we propose two quantization techniques based on K-means clustering and JPEG compression. Numerical experiments with benchmark images illustrate the tradeoffs between communication efficiency, memory use, and reconstruction accuracy.

Miao, Runxuan↗

LC-MEMENTO: A Memory Model for Accelerated Architectures

With the advent of heterogeneous architectures, in particular, with the ubiquity of multi-GPU systems, it is becoming increasingly important to manage device memory efficiently in order to reap the benefits of the additional core count. To date, such responsibility mainly falls on the programmer where device-to-host data communication (and vice versa), if not done properly, may incur costly memory transfer operations and synchronization. The problem may be compounded by additional requirement to maintain system-wide memory consistency that may involve expensive synchronization overhead. In this paper, we present Location Consistency Memory Model for Enhanced Transfer Operations (LC-MEMENTO). This framework considers incorporating runtime techniques for multi-GPU memory management to support relaxed synchronization semantics and memory transfer operations automatically. Specifically, we implement a relaxed form of a memory consistency model based on the Location Consistency (LC) in an Asynchronous Many-Task Runtime (ARTS) and demonstrate that, this memory model enables additional optimization opportunities for the three representative applications encompassing different computational patterns (scientific computation, graphs, data streaming, etc.).

Memory Models, Accelerators, Adaptive Optimization↗

Optimal periodic frequency combs for high-efficiency optical quantum memory based on rare-earth ion crystals

The possibility of enhancing the quantum efficiency of photon echo broadband quantum memory in rare-earth-ion-doped crystals with a frequency comb structure of optical transition inhomogeneous broadening is studied. We have found the optimal parameters of the frequency combs for implementing the pre-assigned quantum efficiency in the chosen spectral range of the optical transitions of rare-earth ions with real parameters. The obtained results allow the conditions to be formulated for increasing the efficiency of broadband quantum memory. The possibilities of experimental implementation of such memory are also discussed. (paper)

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Transverse mode revival of a light-compensated quantum memory

A long-lived quantum memory was developed based on light-compensated cold {sup 87}Rb atoms in a dipole trap. The lifetime of the quantum memory was improved by 40 folds, from 0.67 to 28 ms with the help of a compensation laser beam. Oscillations of the memory efficiency due to the transverse mode breathing of the singly excited spin wave have been clearly observed and clarified with a Monte Carlo simulation procedure. With detailed analysis of the decoherence processes of the spin wave in cold atomic ensembles, this experiment provides a benchmark for the further development of high-quality quantum memories.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

High-Dimensional Similarity Search with Quantum-Assisted Variational Autoencoder

Recent progress in quantum algorithms and hardware indicates the potential importance of quantum computing in the near future. However, finding suitable application areas remains an active area of research. Quantum machine learning is touted as a potential approach to demonstrate quantum advantage within both the gate-model and the adiabatic schemes. For instance, the QVAE has been proposed as a quantum enhancement to the discrete VAE. We extend on previous work and study the real-world applicability of a QVAE by presenting a proof-of-concept for similarity search in large-scale high-dimensional datasets. While exact and fast similarity search algorithms are available for low dimensional datasets, scaling to high-dimensional data is non-trivial. We show how to construct a space-efficient search index based on the latent space representation of a QVAE. Our experiments show a correlation between the Hamming distance in the embedded space and the Euclidean distance in the original space on the MODIS dataset. Further, we find real-world speedups compared to linear search and demonstrate memory-efficient scaling to half a billion data points.

Data mining, similarity search, quantum machine le↗

High dimensional similarity search with quantum assisted variational autoencoder

Recent progress in quantum algorithms and hardware is indicator of the potential importance of quantum computing in the next future. However, finding suitable application areas remains an active area of research. Quantum machine learning [1] is touted as a potential approach to demonstrate quantum advantage within both the gate-model [2,3] and the adiabatic [4,5] schemes. For instance, the Quantum-assisted Variational Autoencoder (QVAE) [6] has been proposed as a quantum enhancement to the discrete VAE [7]. We extend on previous work and study the real-world applicability of a QVAE, specifically, for similarity search in large-scale high dimensional datasets. While similarity search algorithms are available for low dimensional datasets, scaling to billion-scale datasets with thousands of dimensions is non-trivial. We show how the latent-space representation of a QVAE can be used to construct a space-efficient search index. We back up our claims by experimental results which show a correlation between the Hamming distance in the embedded space and the Euclidean distance in the original space on the Moderate Resolution Imaging Spectroradiometer (MODIS) dataset. Further, we show real-world speedups compared to linear search and demonstrate memory efficient scaling to large-scale datasets.

Nicholas D Gao↗

CMS event processing multi-core efficiency status

In 2015, CMS was the first LHC experiment to begin using a multi-threaded framework for doing event processing. This new framework utilizes Intel’s Thread Building Block library to manage concurrency via a task based processing model. During the 2015 LHC run period, CMS only ran reconstruction jobs using multiple threads because only those jobs were sufficiently thread efficient. Recent work now allows simulation and digitization to be thread efficient. In addition, during 2015 the multi-threaded framework could run events in parallel but could only use one thread per event. Work done in 2016 now allows multiple threads to be used while processing one event. In this presentation we will show how these recent changes have improved CMS’s overall threading and memory efficiency and we will discuss work to be done to further increase those efficiencies.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗