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 73 records · Page 4

Performance Analysis of the NAS Y-MP Workload

This paper describes the performance characteristics of the computational workloads on the NAS Cray Y-MP machines, a Y-MP 832 and later a Y-MP 8128. Hardware measurements indicated that the Y-MP workload performance matured over time, ultimately sustaining an average throughput of 0.8 GFLOPS and a vector operation fraction of 87%. The measurements also revealed an operation rate exceeding 1 per clock period, a well-balanced architecture featuring a strong utilization of vector functional units, and an efficient memory organization. Introduction of the larger memory 8128 increased throughput by allowing a more efficient utilization of CPUs. Throughput also depended on the metering of the batch queues; low-idle Saturday workloads required a buffer of small jobs to prevent memory starvation of the CPU. UNICOS required about 7% of total CPU time to service the 832 workloads; this overhead decreased to 5% for the 8128 workloads. While most of the system time went to service I/O requests, efficient scheduling prevented excessive idle due to I/O wait. System measurements disclosed no obvious bottlenecks in the response of the machine and UNICOS to the workloads. In most cases, Cray-provided software tools were- quite sufficient for measuring the performance of both the machine and operating, system.

Bergeron, Robert J.↗

MATAR: A performance portability and productivity implementation of data-oriented design with Kokkos

There is a need for simple, fast, and memory-efficient multidimensional data structures for dense and sparse storage that arise with numerical methods and in software applications. The data structures must perform equally well across multiple computer architectures, including CPUs and GPUs. For this purpose, we developed MATAR, a C++ software library that allows for simple creation and use of intricate data structures that is also portable across disparate architectures using Kokkos. Here, the performance aspect is achieved by forcing contiguous memory layout (or as close to contiguous as possible) for multidimensional and multi-size dense or sparse MATrix and ARray (hence, MATAR) types. Our results show that MATAR has the capability to improve memory utilization, performance, and programmer productivity in scientific computing. This is achieved by fitting more work into the available memory, minimizing memory loads required, and by loading memory in the most efficient order. This document describes the purpose of the work, the implementation of each of the data types, and the resulting performance both in some simple baseline test cases and in an application code.

97 MATHEMATICS AND COMPUTING↗

Quantum Dynamics at Scale: Ultrafast Control of Emergent Functional Materials

Confluence of extreme-scale quantum dynamics simulations (i.e. quantum@scale) and cutting-edge x-ray free-electron laser experiments are revolutionizing materials science. An archetypal example is the exciting concept of using picosecond light pulses to control emergent material properties on demand in atomically-thin layered materials. This paper describes efforts to scale our quantum molecular dynamics engine toward the United States' first exaflop/s computer, under an Aurora Early Science Program project named "Metascalable layered material genome". Key algorithmic and computing techniques incorporated are: (1) globally-scalable and locally-fast solvers within a linear-scaling divide-conquer-recombine algorithmic framework; (2) algebraic `BLASification' of computational kernels; and (3) data alignment and loop restructuring, along with register and cache blocking, for enhanced vectorization and efficient memory access. The resulting weak-scaling parallel efficiency was 0.93 on 131,072 Intel Xeon Phi cores for a 56.6 million atom (or 169 million valence-electron) system, whereas the various code transformations achieved 5-fold speedup. The optimized simulation engine allowed us for the first time to establish a significant effect of substrate on the dynamics of layered material upon electronic excitation.

Algebraic BLASification↗

Characterization of ECRAM materials and devices

As the limits of Moore’s Law approaches, new computing paradigms are developing to breakthrough this bottleneck. One such computer architecture is neuromorphic computing, which models the brain. Electrochemical random-access memory (ECRAM) is a low power and energy efficient memory due to characteristics, such as in-memory compute, ion modulation of the channel conductance, and computation distribution with large scale array integration.

97 MATHEMATICS AND COMPUTING↗

Three-Dimensional Nacelle Aeroacoustics Code With Application to Impedance Education

A three-dimensional nacelle acoustics code that accounts for uniform mean flow and variable surface impedance liners is developed. The code is linked to a commercial version of the NASA-developed General Purpose Solver (for solution of linear systems of equations) in order to obtain the capability to study high frequency waves that may require millions of grid points for resolution. Detailed, single-processor statistics for the performance of the solver in rigid and soft-wall ducts are presented. Over the range of frequencies of current interest in nacelle liner research, noise attenuation levels predicted from the code were in excellent agreement with those predicted from mode theory. The equation solver is memory efficient, requiring only a small fraction of the memory available on modern computers. As an application, the code is combined with an optimization algorithm and used to reduce the impedance spectrum of a ceramic liner. The primary problem with using the code to perform optimization studies at frequencies above I1kHz is the excessive CPU time (a major portion of which is matrix assembly). The research recommends that research be directed toward development of a rapid sparse assembler and exploitation of the multiprocessor capability of the solver to further reduce CPU time.

Watson, Willie R.↗

Streaming Compression of Hexahedral Meshes

We describe a method for streaming compression of hexahedral meshes. Given an interleaved stream of vertices and hexahedral our coder incrementally compresses the mesh in the presented order. Our coder is extremely memory efficient when the input stream documents when vertices are referenced for the last time (i.e. when it contains topological finalization tags). Our coder then continuously releases and reuses data structures that no longer contribute to compressing the remainder of the stream. This means in practice that our coder has only a small fraction of the whole mesh in memory at any time. We can therefore compress very large meshes - even meshes that do not file in memory. Compared to traditional, non-streaming approaches that load the entire mesh and globally reorder it during compression, our algorithm trades a less compact compressed representation for significant gains in speed, memory, and I/O efficiency. For example, on the 456k hexahedra 'blade' mesh, our coder is twice as fast and uses 88 times less memory (only 3.1 MB) with the compressed file increasing about 3% in size. We also present the first scheme for predictive compression of properties associated with hexahedral cells.

99 GENERAL AND MISCELLANEOUS↗

Memorial Pools Energy Efficiency Retrofits

The National September 11 Memorial & Museum completed a retrofit of the submergible LED lighting fixtures in its North and South Memorial Pools, located at the World Trade Center site in New York City. The two one-acre pools are illuminated nightly by custom LED fixtures set into their base, and the original lights — in place since 2011 — had begun reaching the end of their useful life, leaving sections of the pools dim or dark. Rather than fully replacing the fixtures, the project team worked with the original manufacturer, Acuity Brands, to retrofit and reuse existing components across all eight pool walls (34 fixtures per wall), reducing cost and waste. Fixtures were shipped in custom crates to Minnesota for retrofitting, then reinstalled onsite by Memorial & Museum staff and contractor ABM. The team addressed two key technical challenges during the project: early leakage in retrofitted fixtures, resolved by introducing vacuum-sealing and nitrogen-fill testing before shipment; and chord damage during transit, resolved with custom-designed shipping crates. All eight pool walls were successfully retrofitted and reinstalled by August 27, 2025, ahead of the 25th anniversary of the September 11, 2001 attacks. Sitewide electrical consumption in September 2025 reflected an approximate 7% reduction, contributing to the institution's broader net-zero and LEED Gold sustainability goals.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Portable Electromyograph

A portable electronic apparatus records electromyographic (EMG) signals in as many as 16 channels at a sampling rate of 1,024 Hz in each channel. The apparatus (see figure) includes 16 differential EMG electrodes (each electrode corresponding to one channel) with cables and attachment hardware, reference electrodes, an input/output-and-power-adapter unit, a 16-bit analog-to-digital converter, and a hand-held computer that contains a removable 256-MB flash memory card. When all 16 EMG electrodes are in use, full-bandwidth data can be recorded in each channel for as long as 8 hours. The apparatus is powered by a battery and is small enough that it can be carried in a waist pouch. The computer is equipped with a small screen that can be used to display the incoming signals on each channel. Amplitude and time adjustments of this display can be made easily by use of touch buttons on the screen. The user can also set up a data-acquisition schedule to conform to experimental protocols or to manage battery energy and memory efficiently. Once the EMG data have been recorded, the flash memory card is removed from the EMG apparatus and placed in a flash-memory- card-reading external drive unit connected to a personal computer (PC). The PC can then read the data recorded in the 16 channels. Preferably, before further analysis, the data should be stored in the hard drive of the PC. The data files are opened and viewed on the PC by use of special- purpose software. The software for operation of the apparatus resides in a random-access memory (RAM), with backup power supplied by a small internal lithium cell. A backup copy of this software resides on the flash memory card. In the event of loss of both main and backup battery power and consequent loss of this software, the backup copy can be used to restore the RAM copy after power has been restored. Accessories for this device are also available. These include goniometers, accelerometers, foot switches, and force gauges.

De Luca, Gianluca↗

Algorithms for Performance, Dependability, and Performability Evaluation using Stochastic Activity Networks

Modeling tools and technologies are important for aerospace development. At the University of Illinois, we have worked on advancing the state of the art in modeling by Markov reward models in two important areas: reducing the memory necessary to numerically solve systems represented as stochastic activity networks and other stochastic Petri net extensions while still obtaining solutions in a reasonable amount of time, and finding numerically stable and memory-efficient methods to solve for the reward accumulated during a finite mission time. A long standing problem when modeling with high level formalisms such as stochastic activity networks is the so-called state space explosion, where the number of states increases exponentially with size of the high level model. Thus, the corresponding Markov model becomes prohibitively large and solution is constrained by the the size of primary memory. To reduce the memory necessary to numerically solve complex systems, we propose new methods that can tolerate such large state spaces that do not require any special structure in the model (as many other techniques do). First, we develop methods that generate row and columns of the state transition-rate-matrix on-the-fly, eliminating the need to explicitly store the matrix at all. Next, we introduce a new iterative solution method, called modified adaptive Gauss-Seidel, that exhibits locality in its use of data from the state transition-rate-matrix, permitting us to cache portions of the matrix and hence reduce the solution time. Finally, we develop a new memory and computationally efficient technique for Gauss-Seidel based solvers that avoids the need for generating rows of A in order to solve Ax = b. This is a significant performance improvement for on-the-fly methods as well as other recent solution techniques based on Kronecker operators. Taken together, these new results show that one can solve very large models without any special structure.

Deavours, Daniel D.↗

Autonomous Byte Stream Randomizer

Net-centric networking environments are often faced with limited resources and must utilize bandwidth as efficiently as possible. In networking environments that span wide areas, the data transmission has to be efficient without any redundant or exuberant metadata. The Autonomous Byte Stream Randomizer software provides an extra level of security on top of existing data encryption methods. Randomizing the data s byte stream adds an extra layer to existing data protection methods, thus making it harder for an attacker to decrypt protected data. Based on a generated crypto-graphically secure random seed, a random sequence of numbers is used to intelligently and efficiently swap the organization of bytes in data using the unbiased and memory-efficient in-place Fisher-Yates shuffle method. Swapping bytes and reorganizing the crucial structure of the byte data renders the data file unreadable and leaves the data in a deconstructed state. This deconstruction adds an extra level of security requiring the byte stream to be reconstructed with the random seed in order to be readable. Once the data byte stream has been randomized, the software enables the data to be distributed to N nodes in an environment. Each piece of the data in randomized and distributed form is a separate entity unreadable on its own right, but when combined with all N pieces, is able to be reconstructed back to one. Reconstruction requires possession of the key used for randomizing the bytes, leading to the generation of the same cryptographically secure random sequence of numbers used to randomize the data. This software is a cornerstone capability possessing the ability to generate the same cryptographically secure sequence on different machines and time intervals, thus allowing this software to be used more heavily in net-centric environments where data transfer bandwidth is limited.

Paloulian, George K.↗

BCSR on GPU: A Way Forward Extreme-scale Graph Processing on Accelerator-enabled Frontier Supercomputer

Handling large graphs in a distributed environment requires effective partitioning across processors and efficient management of local partitions. In 2D partitioning, local graphs often become too sparse, making memory-efficient data structures crucial. Using the Compressed Sparse Row (CSR) format wastes space, especially for > 83% of vertices with empty edges for the sparse graphs. This study explores bit-CSR (BCSR), a modified CSR representation, on GPUs to reduce memory usage in graph computations. We achieved 16.67% memory savings on a sparse rmat dataset with 268 million vertices and 357 million edges, without performance degradation, supported by both theoretical and experimental storage savings of 33%. However, we observed a 1.7× slowdown in degree lookup times due to bitwise operations on AMD CPUs. This analysis highlights the potential of BCSR on GPUs for improving Graph500 benchmark performance on GPU-accelerated systems, such as the Frontier supercomputer.

Sattar, Naw Safrin↗

A Study of Clustering Techniques and Hierarchical Matrix Formats for Kernel Ridge Regression

We present memory-efficient and scalable algorithms for kernel methods used in machine learning. Using hierarchical matrix approximations for the kernel matrix the memory requirements, the number of floating point operations, and the execution time are drastically reduced compared to standard dense linear algebra routines. We consider both the general H matrix hierarchical format as well as Hierarchically Semi-Separable (HSS) matrices. Furthermore, we investigate the impact of several preprocessing and clustering techniques on the hierarchical matrix compression. Effective clustering of the input leads to a ten-fold increase in efficiency of the compression. The algorithms are implemented using the STRUMPACK solver library. These results confirm that - with correct tuning of the hyperparameters - classification using kernel ridge regression with the compressed matrix does not lose prediction accuracy compared to the exact - not compressed - kernel matrix and that our approach can be extended to O(1M) datasets, for which computation with the full kernel matrix becomes prohibitively expensive. We present numerical experiments in a distributed memory environment up to 1,024 processors of the NERSC's Cori supercomputer using well-known datasets to the machine learning community that range from dimension 8 up to 784.

97 MATHEMATICS AND COMPUTING↗

Implementation of a Binary Neural Network on a Passive Array of Magnetic Tunnel Junctions

The increasing scale of neural networks and their growing application space have produced demand for more energy- and memory-efficient artificial-intelligence-specific hardware. Avenues to mitigate the main issue, the von Neumann bottleneck, include in-memory and near-memory architectures, as well as algorithmic approaches. In this report we leverage the low-power and the inherently binary operation of magnetic tunnel junctions (MTJs) to demonstrate neural network hardware inference based on passive arrays of MTJs. In general, transferring a trained network model to hardware for inference is confronted by degradation in performance due to device-to-device variations, write errors, parasitic resistance, and nonidealities in the substrate. To quantify the effect of these hardware realities, we benchmark 300 unique weight matrix solutions of a two-layer perceptron to classify the Wine dataset for both classification accuracy and write fidelity. Despite device imperfections, we achieve software-equivalent accuracy of up to 95.3% with proper tuning of network parameters in 15 x 15 MTJ arrays having a range of device sizes. The success of this tuning process shows that new metrics are needed to characterize the performance and quality of networks reproduced in mixed signal hardware.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Grouper: A Compact, Streamable Triangle Mesh Data Structure

Here, we present Grouper: an all-in-one compact file format, random-access data structure, and streamable representation for large triangle meshes. Similarly to the recently published SQuad representation, Grouper represents the geometry and connectivity of a mesh by grouping vertices and triangles into fixed-size records, most of which store two adjacent triangles and a shared vertex. Unlike SQuad, however, Grouper interleaves geometry with connectivity and uses a new connectivity representation to ensure that vertices and triangles can be stored in a coherent order that enables memory-efficient sequential stream processing. We also present a linear-time construction algorithm that allows streaming out Grouper meshes using a small memory footprint while preserving the initial ordering of vertices. In this construction, we show how the problem of assigning vertices and triangles to groups reduces to a well-known NP-hard optimization problem, and present a simple yet effective heuristic solution that performs well in practice. Our array-based Grouper representation also doubles as a triangle mesh data structure that allows direct access to vertices and triangles. Storing only about two integer references per triangle-i.e., less than the three vertex references stored with each triangle in a conventional indexed mesh format-Grouper answers both incidence and adjacency queries in amortized constant time. Our compact representation enables data-parallel processing on multicore computers, instant partitioning and fast transmission for distributed processing, as well as efficient out-of-core access. We demonstrate the versatility and performance benefits of Grouper using a suite of example meshes and processing kernels.

97 MATHEMATICS AND COMPUTING↗

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↗

An Efficient Mixed-Mode Representation of Sparse Tensors

The Compressed Sparse Fiber (CSF) representation for sparse tensors is a generalization of the Compressed Sparse Row (CSR) format for sparse matrices. For a tensor with d modes, typical tensor methods such as CANDECOMP/PARAFAC decomposition (CPD) require a sequence of d tensor computations, where efficient memory access with respect to different modes is required for each of them.The straightforward solution is to use d distinct representations of the tensor, with each one being efficient for one of the d computations. However, a d-fold space overhead is often unacceptable in practice, especially with memory-constrained GPUs. In this paper, we present a mixed-mode tensor representation that partitions the tensor’s nonzero elements into disjoint sections, each of which is compressed to create fibers along a different mode. Experimental results on the latest generation of GPU device demonstrate that better performance can be achieved by using the mixed-mode representation, while utilizing only a small fraction of the space required to keep d distinct CSF representations.

Nisa, Israt↗

INDDGO: Integrated Network Decomposition & Dynamic programming for Graph Optimization

It is well-known that dynamic programming algorithms can utilize tree decompositions to provide a way to solve some NP -hard problems on graphs where the complexity is polynomial in the number of nodes and edges in the graph, but exponential in the width of the underlying tree decomposition. However, there has been relatively little computational work done to determine the practical utility of such dynamic programming algorithms. We have developed software to construct tree decompositions using various heuristics and have created a fast, memory-efficient dynamic programming implementation for solving maximum weighted independent set. We describe our software and the algorithms we have implemented, focusing on memory saving techniques for the dynamic programming. We compare the running time and memory usage of our implementation with other techniques for solving maximum weighted independent set, including a commercial integer programming solver and a semi-definite programming solver. Our results indicate that it is possible to solve some instances where the underlying decomposition has width much larger than suggested by the literature. For certain types of problems, our dynamic programming code runs several times faster than these other methods.

Groër, Christopher S.↗

Logical shadow tomography: Efficient estimation of error-mitigated observables

We introduce a technique to estimate error-mitigated expectation values on noisy quantum computers. Our technique performs shadow tomography on a logical state to produce a memory-efficient classical reconstruction of the noisy density matrix. Using efficient classical post-processing, one can mitigate errors by projecting into the codespace as in subspace expansion and taking powers of the density matrix as in virtual distillation. Relative to subspace expansion which requires Ω (2^((n-1)k) samples to estimate a Pauli observable with an [[n; k]] stabilizer code, our technique requires only Ө(2^k) samples. Relative to virtual distillation, our technique can compute powers of the density matrix without implementing additional copies of quantum states the quantum computer. We present numerical results using logical states encoded with up to sixty physical qubits and show fast convergence to error-free expectation values with only 10^5 samples under 1% depolarizing noise.

quantum computing↗