Search NASA⌕ Search

SEARCH · Search NASA

Results for “memory access 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.

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

206 records · Page 12

Subsurface Characterization and Machine Learning Predictions at Brady Hot Springs: Preprint

Subsurface data analysis, reservoir modeling, and machine learning (ML) techniques have been applied to the Brady Hot Springs (BHS) geothermal field in Nevada, USA to further characterize the subsurface and assist with optimizing reservoir management. Hundreds of reservoir simulations have been conducted in TETRAD-G and CMG STARS to explore different injection and production fluid flow rates and allocations and to develop a training data set for ML. This process included simulating the historical injection and production since 1979 and prediction of future performance through 2040. ML networks were created and trained using TensorFlow based on multilayer perceptron (MLP), long short-term memory (LSTM), and convolutional neural network (CNN) architectures. These networks took as input selected flow rates, injection temperatures, and historical field operation data and produced estimates of future production temperatures. This approach was first successfully tested on a simplified single fracture doublet system, followed by the application to the BHS reservoir. Using an initial BHS dataset with 37 simulated scenarios, the trained and validated network predicted the production temperature for 6 production wells with the mean absolute percentage error of less than 8%. In a complementary analysis effort, the principal component analysis applied to 13 BHS geological parameters revealed that vertical fracture permeability shows the strongest correlation with fault density and fault intersection density. A new BHS reservoir model was developed considering the fault intersection density as proxy for permeability. This new reservoir model helps to explore under-exploited zones in the reservoir. A data gathering plan to obtain additional subsurface data was developed; it includes temperature surveying for three idle injection wells, at which the reservoir simulations indicate high bottom-hole temperatures. The collected data assist with calibrating the reservoir model and may lead to converting these wells to producers to access under-exploited zones in the reservoir. Data gathering activities are planned for the first quarter of 2021.

40 EE - Geothermal Technologies Office (EE-4G)↗

ExaHDF5: Delivering Efficient Parallel I/O on Exascale Computing Systems

Scientific applications at exascale generate and analyze massive amounts of data. A critical requirement of these applications is the capability to access and manage this data efficiently on exascale systems. As such, parallel I/O, the key technology enables moving data between compute nodes and storage, faces monumental challenges from new applications, memory, and storage architectures considered in the designs of exascale systems. As the storage hierarchy is expanding to include node-local persistent memory, burst buffers, etc., as well as disk-based storage, data movement among these layers must be efficient. Parallel I/O libraries of the future should be capable of handling file sizes of many terabytes and beyond. In this paper, we describe new capabilities we have developed in Hierarchical Data Format version 5 (HDF5), the most popular parallel I/O library for scientific applications. HDF5 is one of the most used libraries at the leadership computing facilities for performing parallel I/O on existing HPC systems. The state-of-the-art features we describe include: Virtual Object Layer (VOL), Data Elevator, asynchronous I/O, full-featured single-writer and multiple-reader (Full SWMR), and parallel querying. In this paper, we introduce these features, their implementations, and the performance and feature benefits to applications and other libraries.

97 MATHEMATICS AND COMPUTING↗

HARMONY: Large-Scale Architecture Search for Efficient Hybrid Language Models

As large language models scale to trillions of parameters, their computational and memory requirements present critical challenges for efficient training and deployment. While Mixture of Experts (MoE) architectures enable efficient scaling through sparse parameter activation, and state-space models like Mamba offer linear-time complexity, principled methods for combining these paradigms remain undeveloped. We introduce HARMONY (Hybrid Architecture Research for Mamba, Optimized with Neural efficiencY), a multi-objective evolutionary neural architecture search framework for discovering efficient hybrid language models that integrate Transformer attention mechanisms, Mixture-of-Experts routing, and Mamba state-space components. Through large-scale distributed search using 16,384 MI250X GPUs on the Frontier supercomputer, HARMONY explores a comprehensive design space encompassing six attention variants (MHA, MQA, GQA, MLA, SWA, and Mamba-2), variable MoE configurations with both routed and shared experts, and extensive Mamba hyperparameters. Our framework discovers heterogeneous architectures that balance training performance with computational efficiency through multi-objective optimization incorporating latency penalties and fitness-based selection. Analysis of discovered architectures reveals that optimal hybrid designs favor heterogeneous component mixing rather than homogeneous patterns, with Mamba-2 and Multi-Head Latent Attention (MLA) emerging as preferred mechanisms. Discovered architectures demonstrate superior training efficiency: our best configuration achieves a final perplexity of 1.0874 with 2.38B parameters while processing 4,320 tokens/second, outperforming significantly larger manually designed models. Full-scale evaluation shows HARMONY's top architectures achieve better loss trajectories than equivalently-sized models using state-of-the-art configurations including Mixtral, Jamba, and Samba. Additionally, we demonstrate 91% weak scaling efficiency when training discovered 36B-parameter models across 1,024 GPUs. HARMONY is released as an open framework with comprehensive tools for building and training hybrid models using expert-data-pipeline parallelism, democratizing access to automated architecture design for next-generation language models.

Herron, Emily [ORNL] (ORCID:0000000273008172)↗

Approximate Weighted Matching On Emerging Manycore and Multithreaded Architectures

Graph matching is a prototypical combinatorial problem with many applications in computer science and scientific computing, but algorithms for computing optimal matchings are challenging to parallelize. Approximate matching algorithms provide an alternate route for parallelization, and in many contexts compute near-optimal matchings for large-scale graphs. We present sharedmemory parallel implementations for computing half-approximate weighted matching on state-of-the-art multicore (Intel Nehalem and AMD Magny-Cours), manycore (Nvidia Tesla and Nvidia Fermi) and massively multithreaded (Cray XMT) platforms. We provide two implementations: the first implementation uses shared work queues, and is suited to all these platforms; the second implementation is based on dataflow principles, and exploits the architectural features of the Cray XMT. Using a carefully chosen dataset that exhibits characteristics from a wide range of real-world applications, we show scalable performance across different platforms. In particular, for one instance of the input, an R-MAT graph (RMAT-G), we show speedups of: about 32 on 48 cores of an AMD Magny-Cours; 7 on 8 cores of Intel Nehalem; 3 on Nvidia Tesla and 10 on Nvidia Fermi relative to one core of Intel Nehalem; and 60 on 128 processors of Cray XMT. We demonstrate good weak and strong scaling for graphs with up to a billion edges using up to 12, 800 threads. Given the breadth of this work, we focus on simplicity and portability of software rather than excessive fine-tuning for each platform. To the best of our knowledge, this is the first such large-scale study of the half-approximate weighted matching problem on shared-memory platforms. Driven by the critical enabling role of combinatorial algorithms such as matching in scientific computing and the emergence of informatics applications, there is a growing demand to support irregular computations on current and future computing platforms. In this context, we evaluate the capability of emerging multithreaded platforms to tolerate latency induced by irregular memory access patterns, and to support fine-grained parallelism via light-weight synchronization mechanisms. By contrasting the architectural features of these platforms against the Cray XMT, which is specifically designed to support irregular memory-intensive applications, we delineate the impact of these choices on performance.

Halappanavar, Mahantesh↗

EXAGRAPH: Graph and combinatorial methods for enabling exascale applications

Combinatorial algorithms in general and graph algorithms in particular play a critical enabling role in numerous scientific applications. However, the irregular memory access nature of these algorithms makes them one of the hardest algorithmic kernels to implement on parallel systems. With tens of billions of hardware threads and deep memory hierarchies, the exascale computing systems in particular pose extreme challenges in scaling graph algorithms. The codesign center on combinatorial algorithms, ExaGraph, was established to design and develop methods and techniques for efficient implementation of key combinatorial (graph) algorithms chosen from a diverse set of exascale applications. Algebraic and combinatorial methods have a complementary role in the advancement of computational science and engineering, including playing an enabling role on each other. In this paper, we survey the algorithmic and software development activities performed under the auspices of ExaGraph from both a combinatorial and an algebraic perspective. In particular, we detail our recent efforts in porting the algorithms to manycore accelerator (GPU) architectures. We also provide a brief survey of the applications that have benefited from the scalable implementations of different combinatorial algorithms to enable scientific discovery at scale. We believe that several applications will benefit from the algorithmic and software tools developed by the ExaGraph team.

97 MATHEMATICS AND COMPUTING↗

Compressive Classification for TEM-EELS

Acquiring atomic-resolution images of beam sensitive materials or under dynamic conditions (e.g. operando STEM) is very difficult and requires precise dose fractioning to minimize the dose delivered to the sample and acquire the data more rapidly than dynamics present. Typical methods of low-dose STEM imaging include adjusting the beam current, dwell-time, and probe Scanning transmission electron microscopy (STEM) is used routinely to obtain atomic-resolution images of a small selection of radiation tolerant materials under stable conditions.size. All three of these determine the dose over a spatial area in STEM. Recently, compressive sensing [1] has been applied to electron microscopy. The computational imaging method of inpainting [2] has been proposed as a method to reduce dose and increase acquisition speed for STEM. Inpainting is a technique to recover missing pixels in an image. By skipping pixels (randomly) during acquisition, a fourth dose-control mechanism is realized [3]. A few implementations of randomly-subsampled STEM exist (e.g., [4]). To make the concepts of compressive sensing accessible to the microscopy community we begin with an example using atomic resolution images of a GaAs crystal (see Figure 1). Because the crystal is periodic it has a sparse representation in reciprocal space. In fact, a sparse representation is a key component of compressive sensing. If an image is sub-sampled randomly we find that the Fourier transform remains sparse, but becomes increasing noisy as less and less data is acquired. Yet, by finding the peaks in the reciprocal image and removing the noise it is possible to recover atomic resolution images with only 1% of the pixels. Figure 1 shows the reconstruction process for GaAs sub-sampled at 50%, 10%, 5%, and 1%. Periodic images form a special class for which the optimal sampling approach (according to CS theory) is a random selection of pixels. For non-periodic images, such as those arising in in-situ STEM, methods such as dictionary learning are necessary [2,5]. The data can be compressed in time for TEM [5] or pixels can be skipped randomly for STEM. There are several issues that arise in developing a random scan for STEM [4] and a coded aperture for TEM. The remainder of the talk will address these issues. References: [1] E Candes and M Wakin. Signal Processing Magazine, IEEE. 2(25) (2008), pp. 21. [2] A Stevens, Pu Y, Sun Y, et al. Artificial Intelligence & Statistics (AISTATS), 2017. [3] A Stevens, H Yang, L Carin et al. Microscopy 63(1), (2014), pp. 41. [4] L Kovarik, A Stevens, A Liyu et al. Appl. Phys. Lett. 109, 164102 (2016). [5] A Stevens, L Kovarik, P Abellan et al. Adv. Structural and Chemical Imaging 1(10), (2015), pp. 1. [6] Supported by the Chemical Imaging, Signature Discovery, and Analytics in Motion Initiatives at PNNL. PNNL is operated by Battelle Memorial Inst. for the US DOE; contract DE-AC05-76RL01830.

Hao, Weituo↗

Technical note: Optimizing the in situ cosmogenic 36 Cl extraction and measurement workflow for geologic applications

Abstract. In situ cosmogenic 36Cl analysis by accelerator mass spectrometry (AMS) is routinely employed to date Quaternary surfaces and assess rates of landscape evolution. However, standard laboratory preparation procedures for 36Cl dating require the addition of large amounts of isotopically enriched chlorine spike solution; these solutions are expensive and increasingly difficult to acquire from commercial sources. In addition, the typical workflow for 36Cl dating involves measuring both 35Cl/37Cl and 36Cl/Cl concurrently on the high-energy (post-accelerator) end of the AMS system, but 35Cl/37Cl determinations using this technique can be complicated by isotope fractionation and system memory during measurement. The traditional workflow also does not provide 36Cl extraction laboratories with the data needed to calculate native Cl concentrations in advance of 36Cl/Cl measurements. In light of these concerns, we present an improved workflow for extracting and measuring chlorine in geologic materials. Our initial step is to characterize 35Cl/37Cl on sample aliquots of up to ∼1 g prepared in Ag(Cl, Br) matrices, which greatly reduces the amount of isotopically enriched spike solution required to measure native Cl content in each sample. To avoid potential issues with isotope fractionation through the accelerator, 35Cl/37Cl is measured on the low-energy, pre-accelerator end of the AMS line. Then, for 36Cl/Cl measurements, we extract Cl as AgCl or Ag(Cl, Br) in analytical batches with a consistent total Cl load across all samples; this step is intended to minimize source memory effects during 36Cl/Cl measurements and allows the preparation of AMS standards that are customized to match known Cl contents in the samples. To assess the efficacy of this extraction and measurement workflow, we compare chlorine isotope ratio measurements on seven geologic samples prepared using standard procedures and the updated workflow. Measurements of 35Cl/37Cl and 36Cl/Cl are consistent between the two workflows, and 35Cl/37Cl values measured using our methods have considerably higher precision than those measured following standard protocols. The chemical preparation and measurement workflow presented here (1) reduces the amount of isotopically enriched chlorine spike used per rock sample by up to 95 %; (2) identifies rocks with high native Cl concentrations, which may be lower priority for 36Cl surface exposure dating, at an early stage of analysis; and (3) allows laboratory users to maintain control over the total chlorine content within and across analytical batches. These methods can be incorporated into existing laboratory and AMS protocols for 36Cl analyses and will increase the accessibility of 36Cl dating for geologic applications.

58 GEOSCIENCES↗

Evolving T-cell vaccine strategies for HIV, the virus with a thousand faces

HIV's rapid global spread and the human suffering it has left in its wake have made AIDS a global heath priority for the 25 years since its discovery. Yet its capacity to rapidly evolve has made combating this virus a tremendous challenge. The obstacles to creating an effective HIV vaccine are formidable, but there are advances in the field on many fronts, in terms of novel vectors, adjuvants, and antigen design strategies. SIV live attenuated vaccine models are able to confer protection against heterologous challenge, and this continues to provide opportunities to explore the biological underpinnings of a protective effect (9). More indirect, but equally important, is new understanding regarding the biology of acute infection (43), the role of immune response in long-term non-progression (6,62, 81), and defining characteristics of broadly neutralizing antibodies (4). In this review we will focus on summarizing strategies directed towards a single issue, that of contending with HIV variation in terms of designing aT-cell vaccine. The strategies that prove most effective in this area can ultimately be combined with the best strategies under development in other areas, with the hope of ultimately converging on a viable vaccine candidate. Only two large HIV vaccine efficacy trials have been completed and both have failed to prevent infection or confer a benefit to infected individual (23,34), but there is ample reason to continue our efforts. A historic breakthrough came in 1996, when it was realized that although the virus could escape from a single antiretroviral (ARV) therapy, it could be thwarted by a combination of medications that simultaneously targeted different parts of the virus (HAART) (38). This revelation came after 15 years of research, thought, and clinical testing; to enable that vital progress the research and clinical communities had to first define and understand, then develop a strategy to counter, the remarkable evolutionary potential of the virus. HAART, for the first time, provided an effective treatment to help those with living with HIV stay healthy. Nonetheless, the treatment has limitations. People with HIV face a lifetime of expensive daily multi-drug regimens, often with side effects; drug resistance at the individual and population level are issues (56); and universal access, despite substantial progress, is a dream not yet realized for many of the millions of the world's poor who are living with HIV (68). These issues, combined with the growing numbers of people infected globally and impact of HIV on society, make the development of an HIV vaccine or a prophylactic prevention strategy a crucial if elusive goal. In some ways, the history of HIV vaccine deVelopment has paralleled the early stages of designing effective therapy. We had to test the simple strategies first, but meanwhile the story of the impact of diversity from an immunological perspective is still unfolding, and novel ideas countermeasures are being explored.

59 BASIC BIOLOGICAL SCIENCES↗