SEARCH · Search NASA

Results for “Memory Optimization”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 253 records · Page 14

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Enhancing quantum memory lifetime with measurement-free local error correction and reinforcement learning

Reliable quantum computation requires systematic identification and correction of errors that occur and accumulate in quantum hardware. To diagnose and correct such errors, standard quantum error-correcting protocols utilize global error information across the system obtained by mid-circuit readout of ancillary qubits. We investigate circuit-level error-correcting protocols that are measurement-free and based on local error information. Such a local error correction (LEC) circuit consists of faulty multi-qubit gates to perform both syndrome extraction and ancilla-controlled error removal. We develop and implement a reinforcement learning framework that takes a fixed set of faulty gates as inputs and outputs an optimized LEC circuit. To evaluate this approach, we quantitatively characterize an extension of logical qubit lifetime by a noisy LEC circuit. For the two-dimensional (2D) classical Ising model and four-dimensional toric code, our optimized LEC circuit performs better at extending a memory lifetime compared with a conventional LEC circuit based on Toom's rule in a subthreshold gate error regime. We further show that such circuits can be used to reduce the rate of mid-circuit readouts to preserve a 2D toric code memory. Lastly, we discuss the application of the LEC protocol on dissipative preparation of quantum states with topological phases.

74 ATOMIC AND MOLECULAR PHYSICS↗

DOE OSTI2026

Observable optimization for precision theory: machine learning energy correlators

The practice of collider physics typically involves the marginalization of multi-dimensional collider data to uni-dimensional observables relevant for some physics task. In many cases, such as classification or anomaly detection, the observable can be arbitrarily complicated, such as the output of a neural network. However, for precision measurements, the observable must correspond to something computable systematically beyond the level of current simulation tools. In this work, we demonstrate that precision-theory-compatible observable space exploration can be systematized by using neural simulation-based inference techniques from machine learning. We illustrate this approach by exploring the space of marginalizations of the energy 3-point correlator to optimize sensitivity to the top quark mass. We first learn the energy-weighted probability density from simulation, then search in the space of marginalizations for an optimal triangle shape. Although simulations and machine learning are used in the process of observable optimization, the output is an observable definition which can be then computed to high precision and compared directly to data without any memory of the computations which produced it. We find that the optimal marginalization is isosceles triangles on the sphere with a side ratio approximately $1 : 1 : \sqrt{2}$ (i.e. right triangles) within the set of marginalizations we consider.

Jets and Jet Substructure↗

DOE OSTI2017

Fusion PIC code performance analysis on the Cori KNL system

We study the attainable performance of Particle-In-Cell codes on the Cori KNL system by analyzing a miniature particle push application based on the fusion PIC code XGC1. We start from the most basic building blocks of a PIC code and build up the complexity to identify the kernels that cost the most in performance and focus optimization efforts there. Particle push kernels operate at high AI and are not likely to be memory bandwidth or even cache bandwidth bound on KNL. Therefore, we see only minor benefits from the high bandwidth memory available on KNL, and achieving good vectorization is shown to be the most beneficial optimization path with theoretical yield of up to 8x speedup on KNL. In practice we are able to obtain up to a 4x gain from vectorization due to limitations set by the data layout and memory latency.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

A GPU-based compressible combustion solver for applications exhibiting disparate space and time scales

High-speed chemically active flows pose significant computational challenges due to their disparate space and time scales, with stiff chemistry often dominating simulation time. While modern scientific computing programs achieve exascale performance by leveraging graphics processing units (GPUs), existing GPU-based compressible combustion solvers face critical limitations in memory management, load balancing, and handling the highly localized nature of chemical reactions. To this end, we present a high-performance compressible reacting flow solver built on the AMReX framework and optimized for multi-GPU settings. Here, our approach addresses three GPU performance bottlenecks: memory access patterns through column-major storage optimization, computational workload variability via a bulk-sparse integration strategy for chemical kinetics, and multi-GPU load distribution for adaptive mesh refinement applications. The solver adapts existing matrix-based chemical kinetics formulations to multi-grid contexts. Using representative combustion applications, including 2D and 3D detonations and a 3D jet-in-crossflow configuration, we demonstrate 1.4–5× performance improvements over initial implementations on an in-house cluster of NVIDIA H100 GPUs, and near-ideal weak scaling on the Frontier supercomputer (Oak Ridge Leadership Computing Facility) with up to 1024 AMD Instinct MI250X GPUs. Roofline analysis reveals substantial improvements in arithmetic intensity for both convection (∼ 10 ×) and chemistry (∼ 4 ×) routines, confirming efficient utilization of GPU memory bandwidth and computational resources.

42 ENGINEERING↗

DOE OSTI2017

Block-Parallel Data Analysis with DIY2

DIY2 is a programming model and runtime for block-parallel analytics on distributed-memory machines. Its main abstraction is block-structured data parallelism: data are decomposed into blocks; blocks are assigned to processing elements (processes or threads); computation is described as iterations over these blocks, and communication between blocks is defined by reusable patterns. By expressing computation in this general form, the DIY2 runtime is free to optimize the movement of blocks between slow and fast memories (disk and flash vs. DRAM) and to concurrently execute blocks residing in memory with multiple threads. This enables the same program to execute in-core, out-of-core, serial, parallel, single-threaded, multithreaded, or combinations thereof. This paper describes the implementation of the main features of the DIY2 programming model and optimizations to improve performance. DIY2 is evaluated on benchmark test cases to establish baseline performance for several common patterns and on larger complete analysis codes running on large-scale HPC machines.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2020

UNITY: Unified Memory and Storage Space

UNITY is a 36-month project focused on providing design and evaluate a new distributed storage paradigm that unifies the traditionally distinct application views of memory- and file-based data storage into a single scalable and resilient environment. The project is a collaboration among Oak Ridge National Laboratory (ORNL, Lead institution), Los Alamos National Labs (LANL), and Georgia Tech (GT). The main contributions of the GT team have been around development of low level systems software for best leveraging the capabilities of new types of persistent memory technologies, and for development of methods for intelligent data management across memory/storage substrates with heterogeneous components. GT contributed the Phoenix library for optimized checkpoint/restart for HPC I/O for systems with non-volatile memory (NVM), the NVStream library for NVM-specialized streaming I/O for HPC workflows, the CoMerge, Mnemo and Kleio solutions for intelligent data management on NVM-based systems. These contributions result in significant improvements in both application performance and system efficiency.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Efficient Training of Deep Neural Operator Networks via Randomized Sampling

Neural operators (NOs) employ deep neural networks to learn the mappings between infinitedimensional function spaces. Deep operator network (DeepONet), a popular NO architecture, has demonstrated success in the real-time prediction of complex dynamics across various scientific and engineering applications. In this work, we introduce a random sampling technique to be adopted during the training of DeepONet, aimed at improving the generalization ability of the model, while significantly reducing the computational time. The proposed approach targets the trunk network of the DeepONet model that outputs the basis functions corresponding to the spatiotemporal locations of the bounded domain on which the physical system is defined. While constructing the loss function, DeepONet training traditionally considers a uniform grid of spatiotemporal points at which all the output functions are evaluated for each iteration. This approach leads to a larger batch size, resulting in poor generalization and increased memory demands, due to the limitations of the stochastic gradient descent (SGD) optimizer. The proposed random sampling over the inputs of the trunk net mitigates these challenges, improving generalization and reducing the memory requirements during training, resulting in significant computational gains. We validate our hypothesis through three benchmark examples, demonstrating substantial reductions in training time while achieving comparable or lower overall test errors relative to the traditional training approach. Our results indicate that incorporating randomization in the trunk network inputs during training enhances the efficiency and robustness of DeepONet, offering a promising avenue for improving the framework’s performance in modeling complex physical systems.

Karumuri, Sharmila [Department of Civil & Systems ↗

DOE OSTI2025

Analog In-Memory Computing for the Synthetic Aperture Radar Polar Format Algorithm

As the utility of synthetic aperture radar (SAR) systems increases in autonomous vehicles, satellites, and other power- and space-constrained edge applications, there is a growing need for processors that can form SAR images at low power. In recent years, analog in-memory compute (AIMC) has shown immense promise for accelerating neural networks and other matrix-vector multiplication (MVM) heavy workloads at the edge. Here, in this work, we examine how the polar format algorithm (PFA), a popular SAR image formation algorithm, can be mapped to these AIMC systems. The PFA maps readily onto analog MVMs because it primarily consists of two linear operations: interpolation of frequency-domain data to a Cartesian grid, followed by a 2-D Fourier transform. This work presents two approaches to map the interpolation operation onto MVMs in analog hardware: a chirp transform and a modified form of sinc interpolation. These mappings introduce algorithmic errors, and their effect on the quality of SAR image formation is examined, both quantitatively and qualitatively. In addition, the impact of errors introduced by the analog hardware is explored to determine which approach is optimal under varying assumptions about the underlying analog memory devices and circuits.

Analog computing↗

DOE OSTI2019

A memory-distributed quasi-Newton solver for nonlinear programming problems with a small number of general constraints

We approach the problem of parallelizing state-of-the-art nonlinear programming optimization algorithms.Specifically, we focus on parallelizing quasi-Newton interior-point methods that use limited-memory secant Hessian approximations. Such interior-point methods are known to have better convergence properties and to be more effective on large-scale problems than gradient-based and derivative-free optimization algorithms. We target nonlinear and potentially nonconvex optimization problems with an arbitrary number of bound constraints and a small number of general equality and inequality constraints on the optimization variables. These problems occur for example in the form of optimal control, optimal design, and inverse problems governed by ordinary or partial differential equations, whenever they are expressed in a “reduced-space” optimization approach. We introduce and analyze the time and space complexity of a decomposition method for solving the quasi-Newton linear systems that leverages the fact that the quasi-Newton Hessian matrix has a small number of dense blocks that border a low-rank update of a diagonal matrix. This enables an efficient parallelization on memory-distributed computers of the iterations of the optimization algorithm, a state-of-the-art filter line-search interior-point algorithm by Wächter et. al. We illustrate the efficiency of the proposed method by solving structural topology optimization problems on up to 4608 cores on a parallel machine.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2020

Optimizing High Performance Markov Clustering for Pre-Exascale Architectures

HipMCL is a high-performance distributed memory implementation of the popular Markov Cluster Algorithm (MCL) and can cluster large-scale networks within hours using a few thousand CPU-equipped nodes. It relies on sparse matrix computations and heavily makes use of the sparse matrix-sparse matrix multiplication kernel (SpGEMM). The existing parallel algorithms in HipMCL are not scalable to Exascale architectures, both due to their communication costs dominating the runtime at large concurrencies and also due to their inability to take advantage of accelerators that are increasingly popular. In this work, we systematically remove scalability and performance bottlenecks of HipMCL. We enable GPUs by performing the expensive expansion phase of the MCL algorithm on GPU. Additionally, we propose a CPU-GPU joint distributed SpGEMM algorithm called pipelined Sparse SUMMA and integrate a probabilistic memory requirement estimator that is fast and accurate. Furthermore, we develop a new merging algorithm for the incremental processing of partial results produced by the GPUs, which improves the overlap efficiency and the peak memory usage. We also integrate a recent and faster algorithm for performing SpGEMM on CPUs. We validate our new algorithms and optimizations with extensive evaluations. With the enabling of the GPUs and integration of new algorithms, HipMCL is up to 12.4x faster, being able to cluster a network with 70 million proteins and 68 billion connections just under 15 minutes using 1024 nodes of ORNL's Summit supercomputer.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2016

PIPS-SBB: A Parallel Distributed-Memory Branch-and-Bound Algorithm for Stochastic Mixed-Integer Programs

Stochastic mixed-integer programs (SMIPs) deal with optimization under uncertainty at many levels of the decision-making process. When solved as extensive formulation mixed- integer programs, problem instances can exceed available memory on a single workstation. In order to overcome this limitation, we present PIPS-SBB: a distributed-memory parallel stochastic MIP solver that takes advantage of parallelism at multiple levels of the optimization process. We also show promising results on the SIPLIB benchmark by combining methods known for accelerating Branch and Bound (B&B) methods with new ideas that leverage the structure of SMIPs. Finally, we expect the performance of PIPS-SBB to improve further as more functionality is added in the future.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2014

Communication-avoiding symmetric-indefinite factorization

We describe and analyze a novel symmetric triangular factorization algorithm. The algorithm is essentially a block version of Aasen's triangular tridiagonalization. It factors a dense symmetric matrix A as the product A=PLTL T P T where P is a permutation matrix, L is lower triangular, and T is block tridiagonal and banded. The algorithm is the first symmetric-indefinite communication-avoiding factorization: it performs an asymptotically optimal amount of communication in a two-level memory hierarchy for almost any cache-line size. Adaptations of the algorithm to parallel computers are likely to be communication efficient as well; one such adaptation has been recently published. As a result, the current paper describes the algorithm, proves that it is numerically stable, and proves that it is communication optimal.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2019

Tuning crystallographic compatibility to enhance shape memory in ceramics

The extraordinary ability of shape-memory alloys to recover after large imposed deformation motivates efforts to transpose these properties onto ceramics, which would enable practical shape-memory properties at high temperatures and in harsh environments. The theory of mechanical compatibility was utilized to predict promising ceramic candidates in the system (Y 0.5 Ta 0.5 O 2 ) 1–x –(Zr 0.5 Hf 0.5 O 2 ) x , 0.6 < x < 0.85. When these compatibility conditions are met, a reduction in thermal hysteresis by a factor of 2.5, a tripling of deformability, and a 75% enhancement in strain recovery within the shape-memory effect was found. These findings reveal that predicting and optimizing the chemical composition of ceramics to attain improved crystallographic compatibility is a powerful tool for enabling and enhancing their deformability that could ultimately lead to a highly reversible oxide ceramic shape-memory material.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2025

Performance Impact and Trade-Offs for Tuning Key Architectural Parameters on CPU+GPU Systems

In this work, we performed an initial design space exploration of an accelerated processing unit (APU)—a hybrid CPU+GPU architecture that integrates both compute units (CUs) and memory into a unified system. This integration aims to reduce data movement, enhance memory locality, and improve energy efficiency by enabling the CPU and GPU to share memory directly. This effort focused on the interplay of key design components—cache line size, the number of CUs, and main memory technology—and the trade-offs of each configuration were analyzed. This paper highlights the various configurations’ impact on memory accesses, data reuse, and power utilization. The results provide valuable insights that can be leveraged to optimize APU architectures for high-performance and energy-efficient computing and thus create a balanced architecture. This optimization can be achieved by adopting dynamic cache management, runtime CU scaling, and advanced memory integration, highlighting the potential of APUs to address critical challenges in compute, data movement, and memory power consumption.

Asifuzzaman, Kazi [ORNL] (ORCID:0000000240044791)↗

DOE OSTI2016

Challenges of Monte Carlo Transport

These are slides from a presentation for Parallel Summer School at Los Alamos National Laboratory. Solving discretized partial differential equations (PDEs) of interest can require a large number of computations. We can identify concurrency to allow parallel solution of discrete PDEs. Simulated particles histories can be used to solve the Boltzmann transport equation. Particle histories are independent in neutral particle transport, making them amenable to parallel computation. Physical parameters and method type determine the data dependencies of particle histories. Data requirements shape parallel algorithms for Monte Carlo. Then, Parallel Computational Physics and Parallel Monte Carlo are discussed and, finally, the results are given. The mesh passing method greatly simplifies the IMC implementation and allows simple load-balancing. Using MPI windows and passive, one-sided RMA further simplifies the implementation by removing target synchronization. The author is very interested in implementations of PGAS that may allow further optimization for one-sided, read-only memory access (e.g. Open SHMEM). The MPICH_RMA_OVER_DMAPP option and library is required to make one-sided messaging scale on Trinitite - Moonlight scales poorly. Interconnect specific libraries or functions are likely necessary to ensure performance. BRANSON has been used to directly compare the current standard method to a proposed method on idealized problems. The mesh passing algorithm performs well on problems that are designed to show the scalability of the particle passing method. BRANSON can now run load-imbalanced, dynamic problems. Potential avenues of improvement in the mesh passing algorithm will be implemented and explored. A suite of test problems that stress DD methods will elucidate a possible path forward for production codes.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2023

A survey of techniques for optimizing transformer inference

Recent years have seen a phenomenal rise in the performance and applications of transformer neural networks. The family of transformer networks, including Bidirectional Encoder Representations from Transformer (BERT), Generative Pretrained Transformer (GPT) and Vision Transformer (ViT), have shown their effectiveness across Natural Language Processing (NLP) and Computer Vision (CV) domains. Transformer-based networks such as ChatGPT have impacted the lives of common men. However, the quest for high predictive performance has led to an exponential increase in transformers' memory and compute footprint. Researchers have proposed techniques to optimize transformer inference at all levels of abstraction. Further, this paper presents a comprehensive survey of techniques for optimizing the inference phase of transformer networks. We survey techniques such as knowledge distillation, pruning, quantization, neural architecture search and lightweight network design at the algorithmic level. We further review hardware-level optimization techniques and the design of novel hardware accelerators for transformers. We summarize the quantitative results on the number of parameters/FLOPs and the accuracy of several models/techniques to showcase the tradeoff exercised by them. We also outline future directions in this rapidly evolving field of research. We believe that this survey will educate both novice and seasoned researchers and also spark a plethora of research efforts in this field.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2021

Apparatus and method for providing workload distribution of threads among multiple compute units

In some examples, thermal aware optimization logic determines a characteristic (e.g., a workload or type) of a wavefront (e.g., multiple threads). For example, the characteristic indicates whether the wavefront is compute intensive, memory intensive, mixed, and/or another type of wavefront. The thermal aware optimization logic determines temperature information for one or more compute units (CUs) in one or more processing cores. The temperature information includes predictive thermal information indicating expected temperatures corresponding to the one or more CUs and historical thermal information indicating current or past thermal temperatures of at least a portion of a graphics processing unit (GPU). The logic selects the one or more compute units to process the plurality of threads based on the determined characteristic and the temperature information. The logic provides instructions to the selected subset of the plurality of CUs to execute the wavefront.

Rao, Karthik↗

DOE OSTI2023

Apparatus and method for providing workload distribution of threads among multiple compute units

Rao, Karthik↗