SEARCH · Search NASA
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.
The need to solve high-complexity problems using large-scale tightly coupled computing (that is, supercomputing) continues to grow. Here, in this article, we address the needs, challenges, and opportunities for supercomputing over the next decade.
Batched sparse linear algebra operations in general, and solvers in particular, have become the major algorithmic development activity and foremost performance engineering effort in the numerical software libraries work on modern hardware with accelerators such as GPUs. Many applications, ECP and non-ECP alike, require simultaneous solutions of many small linear systems of equations that are structurally sparse in one form or another. In order to move towards high hardware utilization levels, it is important to provide these applications with appropriate interface designs to be both functionally efficient and performance portable and give full access to the appropriate batched sparse solvers running on modern hardware accelerators prevalent across DOE supercomputing sites since the inception of ECP. To this end, we present here a summary of recent advances on the interface designs in use by HPC software libraries supporting batched sparse linear algebra and the development of sparse batched kernel codes for solvers and preconditioners. We also address the potential interoperability opportunities to keep the corresponding software portable between the major hardware accelerators from AMD, Intel, and NVIDIA, while maintaining the appropriate disclosure levels conforming to the active NDA agreements. The presented interface specifications include a mix of batched band, sparse iterative, and sparse direct solvers with their accompanying functionality that is already required by the application codes or we anticipated to be needed in the near future. This report summarizes progress in Kokkos Kernels and the xSDK libraries MAGMA, Ginkgo, hypre, PETSc, and SuperLU.
As Lawrence Livermore National Laboratory’s (LLNL’s) Laboratory Directed Research and Development (LDRD) program enters its fifth decade of leading-edge research and development, its impact and importance have never been stronger. The program continues to advance strategic investments in pioneering science, technology, and engineering, ensuring LLNL will be ready to deliver on our mission as it evolves over the coming decades. Investing in LDRD research, and the people who perform this critical work, gives LLNL the ability to sustain our role as a leader in the Department of Energy and National Nuclear Security Administration enterprise. The LDRD program enables high-risk, high-payoff research that anticipates emerging threats and future mission needs. By nurturing the ingenuity of the Lab’s greatest asset, its people, LDRD funding advances not only our research but also grows and nurtures our workforce: engaging future innovators with student mentoring, challenging postdoctoral researchers to apply their skills to support national security, and strengthening the leadership skills of early career staff. This annual report documents how LDRD investments advance LLNL’s science, technology, and engineering across our mission space. To assess LDRD’s impact we track both short and long-term metrics such as peer-reviewed publications, number of students, or professional fellows. In addition to reviewing these metrics, I encourage you to delve deeper into the breadth of science and technology that illustrate the strategic value of this research portfolio. For instance, a recent exploratory research project used advanced manufacturing to construct miniaturized three-dimensional ion traps for a quantum computer with reduced quantum error rates to enable applications that address national security missions and support basic science. Another project has delved into studying detonation by examining deflagration to enhance the safety and security of the nuclear weapons stockpile. LDRD researchers are also deploying AI agents on two of the world’s most powerful supercomputers to automate and accelerate inertial confinement fusion experiments. Other teams are delivering more accurate optical constants to enable improved validation for aluminum to advance atomic and molecular physics models. LDRD-driven discoveries of how metals deform under extreme conditions strengthen our ability to model and design materials for demanding national security environments. National security challenges are increasingly complex and continuously evolving. LDRD focuses our most innovative science and technology on these challenges, ensuring the Laboratory is developing creative, forward-leaning solutions for our nation and the world. The following pages feature highlights of published scientific advances, patents, and honors that stem from LDRD investments. As you read this report, I hope you will understand how these investments position the Laboratory, and our partners, to meet the demands of the decades ahead.
Earth system predictability is challenged by the complexity of environmental dynamics and the multitude of variables involved. Current AI foundation models, although advanced by leveraging large and heterogeneous data, are often constrained by their size and data integration, limiting their effectiveness in addressing the full range of Earth system prediction challenges. To overcome these limitations, we introduce the Oak Ridge Base Foundation Model for Earth System Predictability (ORBIT), an advanced vision transformer model that scales up to 113 billion parameters using a novel hybrid tensor-data orthogonal parallelism technique. As the largest model of its kind, ORBIT surpasses the current climate AI foundation model size by a thousandfold. Performance scaling tests conducted on the Frontier supercomputer have demonstrated that ORBIT achieves 684 petaFLOPS to 1.6 exaFLOPS sustained throughput, with scaling efficiency maintained at 41% to 85% across 49,152 AMD GPUs. These breakthroughs establish new advances in AI-driven climate modeling and demonstrate promise to significantly improve the Earth system predictability.
Many areas of science exhibit physical processes that are described by high dimensional partial differential equations (PDEs), e.g., the 4D, 5D and 6D models describing magnetized fusion plasmas, models describing quantum chemistry, or derivatives pricing. Such problems are affected by the so-called “curse of dimensionality” where the number of degrees of freedom (or unknowns) required to be solved for scales as N D where N is the number of grid points in any given dimension D. A simple, albeit naive, 6D example is demonstrated in the left panel of Figure 1. With N = 1000 grid points in each dimension, the memory required just to store the solution vector, not to mention forming the matrix required to advance such a system in time, would exceed an exabyte - and also the available memory on the largest of supercomputers available today. The right panel of Figure 1 demonstrates potential savings for a range of problem dimensionalities and grid resolution. While there are methods to simulate such high-dimensional systems, they are mostly based on Monte-Carlo methods, which rely on a statistical sampling such that the resulting solutions include noise. Since the noise in such methods can only be reduced at a rate proportional to $\sqrt{N_p}$ where N p is the number of Monte-Carlo samples, there is a need for continuum, or grid/mesh-based methods for high-dimensional problems, which both do not suffer from noise and bypass the curse of dimensionality. We present a simulation framework that provides such a method using adaptive sparse grids.
We present a new class of AI models for the detection of quasi-circular, spinning, non-precessing binary black hole mergers whose waveforms include the higher order gravitational wave modes ($\ell$, |m|) = {(2,2), (2,1), (3,3), (3,2), (4,4)}, and mode mixing effects in the $\ell$ = 3, |m| = 2 harmonics. These AI models combine hybrid dilated convolution neural networks to accurately model both short- and long-range temporal sequential information of gravitational waves; and graph neural networks to capture spatial correlations among gravitational wave observatories to consistently describe and identify the presence of a signal in a three detector network encompassing the Advanced LIGO and Virgo detectors. We first trained these spatiotemporal-graph AI models using synthetic noise, using 1.2 million modeled waveforms to densely sample this signal manifold, within 1.7 h using 256 NVIDIA A100 GPUs in the Polaris supercomputer at the Argonne Leadership Computing Facility. This distributed training approach exhibited optimal classification performance, and strong scaling up to 512 NVIDIA A100 GPUs. With these AI ensembles we processed data from a three detector network, and found that an ensemble of 4 AI models achieves state-of-the-art performance for signal detection, and reports two misclassifications for every decade of searched data. We distributed AI inference over 128 GPUs in the Polaris supercomputer and 128 nodes in the Theta supercomputer, and completed the processing of a decade of gravitational wave data from a three detector network within 3.5 h. Finally, we fine-tuned these AI ensembles to process the entire month of February 2020, which is part of the O3b LIGO/Virgo observation run, and found 6 gravitational waves, concurrently identified in Advanced LIGO and Advanced Virgo data, and zero false positives. This analysis was completed in one hour using one NVIDIA A100 GPU.
Agentic Retrieval-Augmented Code Synthesis with Iterative RefinementIn supercomputing, efficient and optimized code generation is essential to leverage high-performance systems effectively. We have developed Agentic Retrieval-Augmented Code Synthesis (ARCS), an advanced framework for accurate, robust, and efficient code generation, completion, and translation. ARCS integrates Retrieval-Augmented Generation (RAG) with Chain-of-Thought (CoT) reasoning to systematically break down and iteratively refine complex programming tasks. An agent-based RAG mechanism retrieves relevant code snippets, while real-time execution feedback drives the synthesis of candidate solutions. This process is formalized as a state-action search tree optimization, balancing code correctness with editing efficiency. Evaluations on the Geeks4Geeks and HumanEval benchmarks demonstrate that ARCS significantly outperforms traditional prompting methods in translation and generation quality. By enabling scalable and precise code synthesis, ARCS offers transformative potential for automating and optimizing code development in supercomputing applications, enhancing computational resource utilization
We examine large-eddy-simulation modeling approaches and computational performance of two open-source computational fluid dynamics codes for the simulation of atmospheric boundary layer flows that are of direct relevance to wind energy production. The first code, NekRS, is a high-order, unstructured-grid, spectral element code. The second code, AMR-Wind, is a second-order, block-structured, finite-volume code with adaptive mesh refinement capabilities. The objective of this study is to co-develop these codes in order to improve model fidelity and performance for each. These features will be critical for running ABL-based applications such as wind farm analysis on advanced computing architectures. To this end, we investigate the performance of NekRS and AMR-Wind on the Oak Ridge Leadership Facility supercomputers Summit, using 4 to 800 nodes (24 to 4,800 NVIDIA V100 GPUs), and Crusher, the testbed for the Frontier exascale system, using 18 to 384 Graphics Compute Dies on AMD MI250X GPUs. We compare strong- and weak-scaling capabilities, linear solver performance, and time to solution. We also identify leading inhibitors to parallel scaling.
Tile drainage removes excess water and is an essential, widely adopted management practice to enhance crop productivity in the US Midwest and throughout the world. Tile drainage has been shown to significantly change hydrological and biogeochemical cycles by lowering the water table and reducing the residence time of soil water, although examining the complex interactions and feedbacks in an integrated hydrology–biogeochemistry–crop system remains elusive. Oxygen dynamics are critical to unraveling these interactions and have been ignored or oversimplified in existing models. Understanding these impacts is essential, particularly so because tile drainage has been highlighted as an adaptation under projected wetter springs and drier summers in the changing climate in the US Midwest. We used the ecosys model that uniquely incorporates first-principle soil oxygen dynamics and crop oxygen uptake mechanisms to quantify the impacts of tile drainage on hydrological and biogeochemical cycles and crop growth in corn–soybean rotation fields. The model was validated with data from a multi-treatment, multi-year experiment in Washington, IA. The relative root mean square error (rRMSE) for the corn and soybean yield in validation is 5.66 % and 12.57 %, respectively. The Pearson coefficient (r) of the monthly tile flow during the growing season is 0.78. Plant oxygen stress turns out as an emergent property of the equilibrium between the soil oxygen supply and biological demand. The impact of tile drainage on the system is achieved through a series of coupled feedback mechanisms. The model results show that tile drainage reduces the soil water content and enhances soil oxygenation. It additionally increases the subsurface discharge and elevates inorganic nitrogen leaching, with seasonal variations influenced by climate and crop phenology. The improved aerobic condition alleviates crop oxygen stress during wet springs, thereby promoting crop root growth during the early growth stage. The development of greater root density, in turn, mitigates water stress during dry summers, leading to an overall increase in the crop yield by ∼6 %. These functions indicate the potential of tile drainage in bolstering crop resilience to climate change and the use of this modeling tool for large-scale assessments of tile drainage. The model reveals the underlying causal mechanisms that drive the agroecosystem response to drainage on the coupled hydrology, biogeochemistry, and crop system dynamics.
The Lustre filesystem serves as a vital element in high-performance parallel storage, meeting the rising demands of scientific, research, and enterprise environments. Widely deployed across HPC environments, ranging from small-scale applications in AI/ML, to domains like oil and gas, drug discovery, and meteorology, and manufacturing, Lustre addresses the universal challenge of efficiently accessing vast and ever-increasing volumes of data. Lustre is the filesystem of choice on six out of the top 10 fastest supercomputers in the world today, over 65% of the top 100, and also for over 60% of the top 500. Despite its widespread popularity, there is a lack of a complete and up-to-date reference, covering Lustre’s evolution, design, and various advancements made over the years. In this journal, we aim to fill this gap by providing a comprehensive journey of Lustre, including its history with significant contributions to HPC, detailed architecture and design elements, exploration of advancements added through its evolution, and future directions. Additionally, we present a comparison of Lustre with other prominent storage technologies of the era. To illustrate the current state of Lustre, we analyze several filesystem trends, including utilization, performance, and usage patterns on Orion, the Lustre filesystem on the first exascale supercomputer Frontier. We hope that this journal serves as a comprehensive educational reference for the current and future generations interested in HPC filesystem storage aspects.
Vision-based scientific foundation models hold significant promise for advancing scientific discovery and innovation. This potential stems from their ability to aggregate images from diverse sources—such as varying physical groundings or data acquisition systems—and to learn spatio-temporal correlations using transformer architectures. However, tokenizing and aggregating images can be compute-intensive, a challenge not fully addressed by current distributed methods. In this work, we introduce the Distributed Cross-Channel Hierarchical Aggregation (D-CHAG) approach designed for datasets with a large number of channels across image modalities. Our method is compatible with any model-parallel strategy and any type of vision transformer architecture, significantly improving computational efficiency. We evaluated D-CHAG on hyperspectral imaging and weather forecasting tasks. When integrated with tensor parallelism and model sharding, our approach achieved up to a 75% reduction in memory usage and more than doubled sustained throughput on up to 1,024 AMD GPUs on the Frontier Supercomputer.
The prediction of protein 3D structure from amino acid sequence is a computational grand challenge in biophysics and plays a key role in robust protein structure prediction algorithms, from drug discovery to genome interpretation. The advent of AI models, such as AlphaFold, is revolutionizing applications that depend on robust protein structure prediction algorithms. To maximize the impact, and ease the usability, of these AI tools we introduce APACE, AlphaFold2 and advanced computing as a service, a computational framework that effectively handles this AI model and its TB-size database to conduct accelerated protein structure prediction analyses in modern supercomputing environments. We deployed APACE in the Delta and Polaris supercomputers and quantified its performance for accurate protein structure predictions using four exemplar proteins: 6AWO, 6OAN, 7MEZ, and 6D6U. Using up to 300 ensembles, distributed across 200 NVIDIA A100 GPUs, we found that APACE is up to two orders of magnitude faster than off-the-self AlphaFold2 implementations, reducing time-to-solution from weeks to minutes. This computational approach may be readily linked with robotics laboratories to automate and accelerate scientific discovery.
Abstract The Protein Data Bank (PDB), established in 1971, is the primary global, open‐access archive for experimentally determined 3D macromolecular structures (proteins, RNA, DNA). The research‐focused RCSB.org web‐portal provides access to these data alongside more than one million machine‐learning‐predicted structure models, greatly expanding the available structural landscape. Rapid growth of both experimental and computational structures has increased the need for powerful yet accessible search tools that serve a broad and diverse scientific community. Herein, we describe a redesigned RCSB Protein Data Bank RCSB.org Advanced Search capability that supports intuitive discovery of 3D structures through a unified interface. This interface integrates annotation‐, sequence‐, and 3D structure‐based searches, embeds an interactive 3D viewer, and incorporates curated biological knowledge, such as catalytic site definitions from Mechanism and Catalytic Site Atlas and ligand‐guided structural motifs, for constructing geometry‐driven queries. A new Chemical Search tool allows definition of chemical queries via an integrated drawing tool or standard identifiers, seamlessly combining them with annotation filters. By allowing query definition directly within spatial and chemical contexts, these search interfaces reduce the need for detailed knowledge of residue numbering, chain identifiers, or external cheminformatics software. This capability enables efficient exploration of structures, chemical diversity, and structure–function relationships across all life domains. The redesigned interfaces can be accessed directly at rcsb.org/search/advanced for Advanced Search and rcsb.org/search/chemical for Chemical Search.
The small interaction cross-section of neutrinos makes experimental neutrino physics particularly responsive to technological advancements. A significant development leveraged by the NOvA experiment is large-scale parallel processing, enabling novel computational approaches to longstanding experimental challenges. Central to managing the resulting high-throughput data is NOvA’s implementation of the Freight Train model, designed for efficient data production and handling.This dissertation details the methodology and execution of the NOvA 2024 3-Flavor Oscillation Analysis, supported by a comprehensive dataset spanning ten years. It emphasizes frequentist results refined through the Feldman-Cousins (FC) technique, specifically addressing confidence interval corrections in parameter estimation. The computational intensity associated with Feldman-Cousins arises from extensive Monte Carlo simulations, which were substantially mitigated through parallel computing on the Perlmutter supercomputer at the National Energy Research Scientific Computing Center (NERSC), employing the MPI framework.To further enhance computational efficiency, an Importance Sampling method is introduced and evaluated, demonstrating significant potential to reduce complexity, particularly in exploring extreme parameter space regions. This thesis presents both the successful application of advanced computational resources and the development of sophisticated statistical techniques, aiming to enhance the precision and scope of neutrino oscillation analyses.
We provide the ensemble of fifteen pre-trained graph foundation models (GFMs) for atomistic materials modeling applications. Each one of the fifteen GFMs has been trained on five open-source datasets that (once aggregated) amount to over 154 million atomistic structures, which cover over two-thirds of the natural elements of the periodic table and that comprises a broad set of organic and inorganic compounds. This vast set of atomistic structures comprises ground state configurations that are dynamically stable (i.e., equilibrated structures with atomic forces approximately close to zero values) as well as dynamically unstable structures (i.e., non-equilibrium structures with non-negligible non-zero values of atomic forces). The ensemble of datasets aggregated does NOT include excited states. The datasets have been curated to remove atomistic structures with spectral norm of the force tensor above 100 eV/angstrom. Moreover, a linear term of the energy was computed for each dataset using a linear regression model that uses the chemical concentration of each natural element as regressor. The linear term predicted by the linear regression model has been subtracted from each original energy value to perform a re-alignment of the energy values across different electronic structures approximation theories performed to generate the diverse multi-source, multi-fidelity datasets. The folder "ADIOS_files" contains the set of pre-processed datasets in Adaptable I/O System (ADIOS) format (https://www.exascaleproject.org/research-project/adios/) that have been used for the development and training of GFMs in this work. The "ADIOS_files" directory contains 6 sub-directories named as follows: - ANI1x-v3.bp - MPTrj-v3.bp - OC2020-20M-v3.bp - OC2020-v3.bp - OC2022-v3.bp - qm7x-v3.bp Each sub-directory contains the pre-processed datasets converted in Adaptable I/O System (ADIOS) format (https://www.exascaleproject.org/research-project/adios/) that have been used to the development, training, and performance testing of the ensemble go predictive graph foundation models. Each GFM was developed using HydraGNN (https://github.com/ORNL/HydraGNN) as underlying graph neural network (GNN) architecture. The multi-task learning (MTL) capability of HydraGNN was used to simultaneously train the GFMs on labeled values for direct predictions of energy (a total system property of an atomistic structure that measures the chemical stability) and atomic forces (an atomic level property of an atomistic structure that measures the dynamical stability). The hyper parameters of the GFM have been tuned using scalable hyperparameter optimization (HPO) algorithms implemented in the software DeepHyper (https://github.com/deephyper/deephyper). The pre-training of each HPO trial was performed using distributed data parallelism (DDP) to scale the training across 128 compute nodes of the exascale OLCF supercomputer Frontier. Each HPO trial was trained only for 10 epochs and an early stopping was performed to avoid wasting significant computational resources on GNN architectures that were clearly underperforming. For each HPO trial, the 'omnistat' tool developed by (AMD Research - Advanced Micro Device) was used to measure the total energy consumption in kWh. The ensemble of GFMs was obtained by selecting the fifteen best performing HPO trials. Four models have been selected for their clear advantage in accuracy, and these are the GFMs with IDs 229, 156, 147, 260. Additional eleven models have been selected based on judicious balance between accuracy and energy consumption needed for training, and these are the GFMs with IDs 165, 78, 137, 1, 175, 171, 181, 67, 179, 167, 351. Each selected GFM of the ensemble was continued to cumulate a total of at most 30 epochs. In some cases, the total number of epochs actually performed was les than 30 due to two combined factors: (1) the size of the GFM (i.e., the number of model parameters to train) and (2) the total wall-clock time for which the computational resources could be allocated on OLCF-Frontier. The "Ensemble_of_models" directory contains 15 sub-directories named as follows: - gfm_0.229 - gfm_0.156 - gfm_0.147 - gfm_0.260 - gfm_0.165 - gfm_0.78 - gfm_0.137 - gfm_0.1 - gfm_0.175 - gfm_0.171 - gfm_0.181 - gfm_0.67 - gfm_0.179 - gfm_0.167 - gfm_0.351 Each one of these sub-directories refers to one of the fifteen HPO trials that have been selected to continue the pre-training with at most 30 epochs. With each sub-directory associated with a specific HPO trial, the following files can be found: - config.json: file for argument parsing to develop and train an HydraGNN architecture - gfm_0.ID_epoch_N.pk: file with model parameters for HPO ID trial after N epochs of training The ensemble of fifteen GFM architectures was used for (1) ensemble averaging to stabilize the predictions of energy and atomic forces after pre-training for post-processing analysis and (2) ensemble uncertainty quantification (UQ). The code used to develop, pre-train, and load the pre-trained models for post-processing analysis is available on the ORNL-GitHub at the following link: https://github.com/ORNL/HydraGNN/tree/Predictive_GFM_2024
The Simplified Cloud-Resolving E3SM Atmosphere Model (SCREAM) is the newest addition to the family of Earth System Models capable of explicitly resolving convective systems. SCREAM is a kilometer-scale configuration of the advanced E3SM Atmosphere Model (EAMxx), designed for heterogeneous systems. While the enhanced accuracy of kilometer-scale modeling offers significant benefits, it comes with a substantial computational cost, limiting feasible simulation durations to only a few years, even on the fastest supercomputers. Machine learning presents an opportunity for scientists to achieve the high accuracy of storm-resolving models at a significantly reduced cost. Building on the previous success of applying corrective machine learning (ML) to the FV3 model, this study explores the effects of implementing corrective ML in EAMxx-SCREAM. We also address the computational challenges of integrating the corrective ML, which is written in Python, with the C++/Kokkos EAMxx driver, as well as the potential pitfalls of generalizing an approach that was effective with one atmosphere model to another.
In fiscal year (FY) 2024, the National Renewable Energy Laboratory (NREL) took a major leap forward with the completed full buildout of Kestrel, the Office of Energy Efficiency and Renewable Energy's newest high-performance computing (HPC) system. Kestrel is already supporting science across the portfolio, bringing roughly 44 petaflops of computing power, which is more than five times the capacity of our previous supercomputer, Eagle. By delivering greater GPU capacity, Kestrel enables faster progress in artificial intelligence (AI) and opens new avenues in energy research - from defining long-term planning scenarios to accommodate a growing power system to material discovery to improving energy efficiency in photovoltaics (PV). Across the portfolio, research is being accelerated by Kestrel's impressive power. During FY24, 427 projects and more than 700 researchers used NREL's HPC, supporting the U.S. Department of Energy's Office of Energy Efficiency and Renewable Energy across 13 funding areas. Through these collaborations, researchers produced more than 450 technical outputs, including 195 articles in peer-reviewed publications, pushing the boundaries of science and engineering. This year's report features new sections spotlighting the expanding roles of Artificial Intelligence and Accelerated Computing. We also introduce an early career section to celebrate the accomplishments of our up-and-coming researchers, whose pioneering work is shaping the future of energy. We hope you enjoy the new insights and discoveries highlighted in these pages.
Deep learning has emerged as a powerful method for extracting valuable information from large volumes of data. However, when new training data arrives continuously (i.e., is not fully available from the beginning), incremental training suffers from catastrophic forgetting (i.e., new patterns are reinforced at the expense of previously acquired knowledge). Training from scratch each time new training data becomes available would result in extremely long training times and massive data accumulation. Rehearsal-based continual learning has shown promise for addressing the catastrophic forgetting challenge, but research to date has not addressed performance and scalability. To fill this gap, we propose an approach based on a distributed rehearsal buffer that efficiently complements data-parallel training on multiple GPUs to achieve high accuracy, short runtime, and scalability. It leverages a set of buffers (local to each GPU) and uses several asynchronous techniques for updating these local buffers in an embarrassingly parallel fashion, all while handling the communication overheads necessary to augment input minibatches using unbiased, global sampling. We further propose a generalization of rehearsal buffers to support both classification and generative learning tasks, as well as more advanced rehearsal strategies (notably Dark Experience Replay, leveraging knowledge distillation). We illustrate this approach with a real-life HPC streaming application from the domain of ptychographic image reconstruction. Furthermore, we run extensive experiments on up to 128 GPUs of the ThetaGPU supercomputer to compare our approach with baselines representative of training-from-scratch (the upper bound in terms of accuracy) and incremental training (the lower bound). Results show that rehearsal-based continual learning achieves a top-5 validation accuracy close to the upper bound, while simultaneously exhibiting a runtime close to the lower bound.