arXiv subjects

David E. Keyes

Publications and source records attributed to David E. Keyes.

At least 19 recordsLinked to original sources

arXiv2026

Cross-Layer Energy Analysis of Multimodal Training on Grace Hopper Superchips

Multimodal deep learning models enable joint learning across heterogeneous data sources, including text, images, and video, but their rapid scaling introduces significant memory and communication bottlenecks. As model sizes and sequence lengths increase, training performance becomes increasingly impacted by data movement rather than computation. Frameworks such as DeepSpeed mitigate these challenges through CPU offloading, activation checkpointing, and communication optimizations. However, these techniques introduce additional system activity, which may affect energy efficiency. Meanwhile, tightly integrated heterogeneous architectures, such as the NVIDIA Grace Hopper (GH200) superchip, provide high-bandwidth CPU-GPU interconnects and unified memory, thereby reducing data transfer overhead. In this work, we present a cross-layer analysis of energy and performance trade-offs in multimodal training on GH200 systems, explicitly characterizing the interactions between application, runtime, and hardware layers. Leveraging high-bandwidth CPU-GPU interconnects, our results show that energy efficiency is primarily governed by data movement and overlap rather than raw compute utilization, and that configurations optimized for runtime are not necessarily optimal for energy. Based on these findings, we distill a set of actionable guidelines for practitioners that demonstrate how to balance offloading strategies, sequence parallelism, and hardware-aware scheduling to achieve energy-efficient training. Our results demonstrate that leveraging high-bandwidth CPU-GPU interconnects enables offloading strategies and sequence parallelism, achieving a strong balance among energy efficiency, runtime performance, and computational throughput, providing practical guidelines for efficient multimodal training on modern heterogeneous systems.

cs.DC↗

arXiv2025

High-Performance Statistical Computing (HPSC): Challenges, Opportunities, and Future Directions

We recognize the emergence of a statistical computing community focused on working with large computing platforms and producing software and applications that exemplify high-performance statistical computing (HPSC). The statistical computing (SC) community develops software that is widely used across disciplines. However, it remains largely absent from the high-performance computing (HPC) landscape, particularly on platforms such as those featured on the Top500 or Green500 lists. Many disciplines already participate in HPC, mostly centered around simulation science, although data-focused efforts under the artificial intelligence (AI) label are gaining popularity. Bridging this gap requires both community adaptation and technical innovation to align statistical methods with modern HPC technologies. We can accelerate progress in fast and scalable statistical applications by building strong connections between the SC and HPC communities. We present a brief history of SC, a vision for how its strengths can contribute to statistical science in the HPC environment (such as HPSC), the challenges that remain, and the opportunities currently available, culminating in a possible roadmap toward a thriving HPSC community.

cs.DC↗

arXiv2025

FlashDP: Private Training Large Language Models with Efficient DP-SGD

As large language models (LLMs) increasingly underpin technological advancements, the privacy of their training data emerges as a critical concern. Differential Privacy (DP) serves as a rigorous mechanism to protect this data, yet its integration via Differentially Private Stochastic Gradient Descent (DP-SGD) introduces substantial challenges, primarily due to the complexities of per-sample gradient clipping. Current explicit methods, such as Opacus, necessitate extensive storage for per-sample gradients, significantly inflating memory requirements. Conversely, implicit methods like GhostClip reduce storage needs by recalculating gradients multiple times, which leads to inefficiencies due to redundant computations. This paper introduces FlashDP, an innovative cache-friendly per-layer DP-SGD that consolidates necessary operations into a single task, calculating gradients only once in a fused manner. This approach not only diminishes memory movement by up to \textbf{50\%} but also cuts down redundant computations by \textbf{20\%}, compared to previous methods. Consequently, FlashDP does not increase memory demands and achieves a \textbf{90\%} throughput compared to the Non-DP method on a four-A100 system during the pre-training of the Llama-13B model, while maintaining parity with standard per-layer clipped DP-SGD in terms of accuracy. These advancements establish FlashDP as a pivotal development for efficient and privacy-preserving training of LLMs. FlashDP's code has been open-sourced in https://github.com/kaustpradalab/flashdp.

cs.LG↗

arXiv2025

RCOMPSs: A Scalable Runtime System for R Code Execution on Manycore Systems

R has become a cornerstone of scientific and statistical computing due to its extensive package ecosystem, expressive syntax, and strong support for reproducible analysis. However, as data sizes and computational demands grow, native R parallelism support remains limited. This paper presents RCOMPSs, a scalable runtime system that enables efficient parallel execution of R applications on multicore and manycore systems. RCOMPSs adopts a dynamic, task-based programming model, allowing users to write code in a sequential style, while the runtime automatically handles asynchronous task execution, dependency tracking, and scheduling across available resources. We present RCOMPSs using three representative data analysis algorithms, i.e., K-nearest neighbors (KNN) classification, K-means clustering, and linear regression and evaluate their performance on two modern HPC systems: KAUST Shaheen-III and Barcelona Supercomputing Center (BSC) MareNostrum 5. Experimental results reveal that RCOMPSs demonstrates both strong and weak scalability on up to 128 cores per node and across 32 nodes. For KNN and K-means, parallel efficiency remains above 70% in most settings, while linear regression maintains acceptable performance under shared and distributed memory configurations despite its deeper task dependencies. Overall, RCOMPSs significantly enhances the parallel capabilities of R with minimal, automated, and runtime-aware user intervention, making it a practical solution for large-scale data analytics in high-performance environments.

cs.DC↗

arXiv2025

Scaled Block Vecchia Approximation for High-Dimensional Gaussian Process Emulation on GPUs

Emulating computationally intensive scientific simulations is crucial for enabling uncertainty quantification, optimization, and informed decision-making at scale. Gaussian Processes (GPs) offer a flexible and data-efficient foundation for statistical emulation, but their poor scalability limits applicability to large datasets. We introduce the Scaled Block Vecchia (SBV) algorithm for distributed GPU-based systems. SBV integrates the Scaled Vecchia approach for anisotropic input scaling with the Block Vecchia (BV) method to reduce computational and memory complexity while leveraging GPU acceleration techniques for efficient linear algebra operations. To the best of our knowledge, this is the first distributed implementation of any Vecchia-based GP variant. Our implementation employs MPI for inter-node parallelism and the MAGMA library for GPU-accelerated batched matrix computations. We demonstrate the scalability and efficiency of the proposed algorithm through experiments on synthetic and real-world workloads, including a 50M point simulation from a respiratory disease model. SBV achieves near-linear scalability on up to 512 A100 and GH200 GPUs, handles 2.56B points, and reduces energy use relative to exact GP solvers, establishing SBV as a scalable and energy-efficient framework for emulating large-scale scientific models on GPU-based distributed systems.

cs.DC↗

arXiv2025

ZO2: Scalable Zeroth-Order Fine-Tuning for Extremely Large Language Models with Limited GPU Memory

Fine-tuning large pre-trained LLMs generally demands extensive GPU memory. Traditional first-order optimizers like SGD encounter substantial difficulties due to increased memory requirements from storing activations and gradients during both the forward and backward phases as the model size expands. Alternatively, zeroth-order (ZO) techniques can compute gradients using just forward operations, eliminating the need to store activations. Furthermore, by leveraging CPU capabilities, it's feasible to enhance both the memory and processing power available to a single GPU. We propose a novel framework, ZO2 (Zeroth-Order Offloading), for efficient zeroth-order fine-tuning of LLMs with only limited GPU memory. Our framework dynamically shifts model parameters between the CPU and GPU as required, optimizing computation flow and maximizing GPU usage by minimizing downtime. This integration of parameter adjustments with ZO's double forward operations reduces unnecessary data movement, enhancing the fine-tuning efficacy. Additionally, our framework supports an innovative low-bit precision approach in AMP mode to streamline data exchanges between the CPU and GPU. Employing this approach allows us to fine-tune extraordinarily large models, such as the OPT-175B with more than 175 billion parameters, on a mere 18GB GPU--achievements beyond the reach of traditional methods. Moreover, our framework achieves these results with almost no additional time overhead and absolutely no accuracy loss compared to standard zeroth-order methods. ZO2's code has been open-sourced in https://github.com/liangyuwang/zo2.

cs.LG↗

arXiv2025

GPU-Accelerated Modified Bessel Function of the Second Kind for Gaussian Processes

Modified Bessel functions of the second kind are widely used in physics, engineering, spatial statistics, and machine learning. Since contemporary scientific applications, including machine learning, rely on GPUs for acceleration, providing robust GPU-hosted implementations of special functions, such as the modified Bessel function, is crucial for performance. Existing implementations of the modified Bessel function of the second kind rely on CPUs and have limited coverage of the full range of values needed in some applications. In this work, we present a robust implementation of the modified Bessel function of the second kind on GPUs, eliminating the dependence on the CPU host. We cover a range of values commonly used in real applications, providing high accuracy compared to common libraries like the GNU Scientific Library (GSL) when referenced to Mathematica as the authority. Our GPU-accelerated approach also demonstrates a 2.68X performance improvement using a single A100 GPU compared to the GSL on 40-core Intel Cascade Lake CPUs. Our implementation is integrated into ExaGeoStat, the HPC framework for Gaussian process modeling, where the modified Bessel function of the second kind is required by the Matern covariance function in generating covariance matrices. We accelerate the matrix generation process in ExaGeoStat by up to 12.62X with four A100 GPUs while maintaining almost the same accuracy for modeling and prediction operations using synthetic and real datasets.

cs.DC↗

arXiv2024

Accelerating AI Performance using Anderson Extrapolation on GPUs

We present a novel approach for accelerating AI performance by leveraging Anderson extrapolation, a vector-to-vector mapping technique based on a window of historical iterations. By identifying the crossover point (Fig. 1) where a mixing penalty is incurred, the method focuses on reducing iterations to convergence, with fewer more compute-intensive but generally cacheable iterations, balancing speed and memory usage with accuracy and algorithmic stability, respectively. We demonstrate significant improvements, in both training and inference, motivated by scalability and efficiency extensions to the realm of high-performance computing (HPC).

cs.LG↗

arXiv2024

Accelerating Mixed-Precision Out-of-Core Cholesky Factorization with Static Task Scheduling

This paper explores the performance optimization of out-of-core (OOC) Cholesky factorization on shared-memory systems equipped with multiple GPUs. We employ fine-grained computational tasks to expose concurrency while creating opportunities to overlap data movement asynchronously with computations, especially when dealing with matrices that cannot fit on the GPU memory. We leverage the directed acyclic graph of the task-based Cholesky factorization and map it onto a static scheduler that promotes data reuse while supporting strategies for reducing data movement with the CPU host when the GPU memory is exhausted. The CPU-GPU interconnect may become the main performance bottleneck as the gap between the GPU execution rate and the traditional PCIe bandwidth continues to widen. While the surface-to-volume effect of compute-bound kernels partially mitigates the overhead of data motion, deploying mixed-precision (MxP) computations exacerbates the throughput discrepancy. Using static task scheduling, we evaluate the performance capabilities of the new ultra-fast NVIDIA chip interconnect technology, codenamed NVLink-C2C, that constitutes the backbone of the NVIDIA Grace Hopper Superchip (GH200), against a new four-precision (FP64/FP32/FP16/FP8) left-looking Cholesky factorization. We report the performance results of a benchmarking campaign on various NVIDIA GPU generations and interconnects. We highlight 20% performance superiority against cuSOLVER on a single GH200 with FP64 while hiding the cost of OOC task-based Cholesky factorization, and we scale almost linearly on four GH200 superships. With MxP enabled, our statically scheduled four-precision tile-based Cholesky factorization scores a 3X performance speedup against its FP64-only counterpart, delivering application-worthy FP64 accuracy when modeling a large-scale geospatial statistical application.

cs.DC↗

arXiv2024

Toward Capturing Genetic Epistasis From Multivariate Genome-Wide Association Studies Using Mixed-Precision Kernel Ridge Regression

We exploit the widening margin in tensor-core performance between [FP64/FP32/FP16/INT8,FP64/FP32/FP16/FP8/INT8] on NVIDIA [Ampere,Hopper] GPUs to boost the performance of output accuracy-preserving mixed-precision computation of Genome-Wide Association Studies (GWAS) of 305K patients from the UK BioBank, the largest-ever GWAS cohort studied for genetic epistasis using a multivariate approach. Tile-centric adaptive-precision linear algebraic techniques motivated by reducing data motion gain enhanced significance with low-precision GPU arithmetic. At the core of Kernel Ridge Regression (KRR) techniques for GWAS lie compute-bound cubic-complexity matrix operations that inhibit scaling to aspirational dimensions of the population, genotypes, and phenotypes. We accelerate KRR matrix generation by redesigning the computation for Euclidean distances to engage INT8 tensor cores while exploiting symmetry.We accelerate solution of the regularized KRR systems by deploying a new four-precision Cholesky-based solver, which, at 1.805 mixed-precision ExaOp/s on a nearly full Alps system, outperforms the state-of-the-art CPU-only REGENIE GWAS software by five orders of magnitude.

q-bio.GN↗

arXiv2024

Boosting Earth System Model Outputs And Saving PetaBytes in their Storage Using Exascale Climate Emulators

We present the design and scalable implementation of an exascale climate emulator for addressing the escalating computational and storage requirements of high-resolution Earth System Model simulations. We utilize the spherical harmonic transform to stochastically model spatio-temporal variations in climate data. This provides tunable spatio-temporal resolution and significantly improves the fidelity and granularity of climate emulation, achieving an ultra-high spatial resolution of 0.034 (approximately 3.5 km) in space. Our emulator, trained on 318 billion hourly temperature data points from a 35-year and 31 billion daily data points from an 83-year global simulation ensemble, generates statistically consistent climate emulations. We extend linear solver software to mixed-precision arithmetic GPUs, applying different precisions within a single solver to adapt to different correlation strengths. The PaRSEC runtime system supports efficient parallel matrix operations by optimizing the dynamic balance between computation, communication, and memory requirements. Our BLAS3-rich code is optimized for systems equipped with four different families and generations of GPUs, scaling well to achieve 0.976 EFlop/s on 9,025 nodes (36,100 AMD MI250X multichip module (MCM) GPUs) of Frontier (nearly full system), 0.739 EFlop/s on 1,936 nodes (7,744 Grace-Hopper Superchips (GH200)) of Alps, 0.243 EFlop/s on 1,024 nodes (4,096 A100 GPUs) of Leonardo, and 0.375 EFlop/s on 3,072 nodes (18,432 V100 GPUs) of Summit.

stat.CO↗

arXiv2024

Parallel Approximations for High-Dimensional Multivariate Normal Probability Computation in Confidence Region Detection Applications

Addressing the statistical challenge of computing the multivariate normal (MVN) probability in high dimensions holds significant potential for enhancing various applications. One common way to compute high-dimensional MVN probabilities is the Separation-of-Variables (SOV) algorithm. This algorithm is known for its high computational complexity of O(n^3) and space complexity of O(n^2), mainly due to a Cholesky factorization operation for an n X n covariance matrix, where $n$ represents the dimensionality of the MVN problem. This work proposes a high-performance computing framework that allows scaling the SOV algorithm and, subsequently, the confidence region detection algorithm. The framework leverages parallel linear algebra algorithms with a task-based programming model to achieve performance scalability in computing process probabilities, especially on large-scale systems. In addition, we enhance our implementation by incorporating Tile Low-Rank (TLR) approximation techniques to reduce algorithmic complexity without compromising the necessary accuracy. To evaluate the performance and accuracy of our framework, we conduct assessments using simulated data and a wind speed dataset. Our proposed implementation effectively handles high-dimensional multivariate normal (MVN) probability computations on shared and distributed-memory systems using finite precision arithmetics and TLR approximation computation. Performance results show a significant speedup of up to 20X in solving the MVN problem using TLR approximation compared to the reference dense solution without sacrificing the application's accuracy. The qualitative results on synthetic and real datasets demonstrate how we maintain high accuracy in detecting confidence regions even when relying on TLR approximation to perform the underlying linear algebra operations.

cs.DC↗

arXiv2024

GPU-Accelerated Vecchia Approximations of Gaussian Processes for Geospatial Data using Batched Matrix Computations

Gaussian processes (GPs) are commonly used for geospatial analysis, but they suffer from high computational complexity when dealing with massive data. For instance, the log-likelihood function required in estimating the statistical model parameters for geospatial data is a computationally intensive procedure that involves computing the inverse of a covariance matrix with size n X n, where n represents the number of geographical locations. As a result, in the literature, studies have shifted towards approximation methods to handle larger values of n effectively while maintaining high accuracy. These methods encompass a range of techniques, including low-rank and sparse approximations. Vecchia approximation is one of the most promising methods to speed up evaluating the log-likelihood function. This study presents a parallel implementation of the Vecchia approximation, utilizing batched matrix computations on contemporary GPUs. The proposed implementation relies on batched linear algebra routines to efficiently execute individual conditional distributions in the Vecchia algorithm. We rely on the KBLAS linear algebra library to perform batched linear algebra operations, reducing the time to solution compared to the state-of-the-art parallel implementation of the likelihood estimation operation in the ExaGeoStat software by up to 700X, 833X, 1380X on 32GB GV100, 80GB A100, and 80GB H100 GPUs, respectively. We also successfully manage larger problem sizes on a single NVIDIA GPU, accommodating up to 1M locations with 80GB A100 and H100 GPUs while maintaining the necessary application accuracy. We further assess the accuracy performance of the implemented algorithm, identifying the optimal settings for the Vecchia approximation algorithm to preserve accuracy on two real geospatial datasets: soil moisture data in the Mississippi Basin area and wind speed data in the Middle East.

stat.CO↗

arXiv2023

Portability and Scalability Evaluation of Large-Scale Statistical Modeling and Prediction Software through HPC-Ready Containers

HPC-based applications often have complex workflows with many software dependencies that hinder their portability on contemporary HPC architectures. In addition, these applications often require extraordinary efforts to deploy and execute at performance potential on new HPC systems, while the users expert in these applications generally have less expertise in HPC and related technologies. This paper provides a dynamic solution that facilitates containerization for transferring HPC software onto diverse parallel systems. The study relies on the HPC Workflow as a Service (HPCWaaS) paradigm proposed by the EuroHPC eFlows4HPC project. It offers to deploy workflows through containers tailored for any of a number of specific HPC systems. Traditional container image creation tools rely on OS system packages compiled for generic architecture families (x86\_64, amd64, ppc64, ...) and specific MPI or GPU runtime library versions. The containerization solution proposed in this paper leverages HPC Builders such as Spack or Easybuild and multi-platform builders such as buildx to create a service for automating the creation of container images for the software specific to each hardware architecture, aiming to sustain the overall performance of the software. We assess the efficiency of our proposed solution for porting the geostatistics ExaGeoStat software on various parallel systems while preserving the computational performance. The results show that the performance of the generated images is comparable with the native execution of the software on the same architectures. On the distributed-memory system, the containerized version can scale up to 256 nodes without impacting performance.

cs.DC↗

arXiv2022

The Second Competition on Spatial Statistics for Large Datasets

In the last few decades, the size of spatial and spatio-temporal datasets in many research areas has rapidly increased with the development of data collection technologies. As a result, classical statistical methods in spatial statistics are facing computational challenges. For example, the kriging predictor in geostatistics becomes prohibitive on traditional hardware architectures for large datasets as it requires high computing power and memory footprint when dealing with large dense matrix operations. Over the years, various approximation methods have been proposed to address such computational issues, however, the community lacks a holistic process to assess their approximation efficiency. To provide a fair assessment, in 2021, we organized the first competition on spatial statistics for large datasets, generated by our {\em ExaGeoStat} software, and asked participants to report the results of estimation and prediction. Thanks to its widely acknowledged success and at the request of many participants, we organized the second competition in 2022 focusing on predictions for more complex spatial and spatio-temporal processes, including univariate nonstationary spatial processes, univariate stationary space-time processes, and bivariate stationary spatial processes. In this paper, we describe in detail the data generation procedure and make the valuable datasets publicly available for a wider adoption. Then, we review the submitted methods from fourteen teams worldwide, analyze the competition outcomes, and assess the performance of each team.

stat.OT↗

arXiv2021

H2Opus: A distributed-memory multi-GPU software package for non-local operators

Hierarchical $\mathcal{H}^2$-matrices are asymptotically optimal representations for the discretizations of non-local operators such as those arising in integral equations or from kernel functions. Their $O(N)$ complexity in both memory and operator application makes them particularly suited for large-scale problems. As a result, there is a need for software that provides support for distributed operations on these matrices to allow large-scale problems to be represented. In this paper, we present high-performance, distributed-memory GPU-accelerated algorithms and implementations for matrix-vector multiplication and matrix recompression of hierarchical matrices in the $\mathcal{H}^2$ format. The algorithms are a new module of H2Opus, a performance-oriented package that supports a broad variety of $\mathcal{H}^2$-matrix operations on CPUs and GPUs. Performance in the distributed GPU setting is achieved by marshaling the tree data of the hierarchical matrix representation to allow batched kernels to be executed on the individual GPUs. MPI is used for inter-process communication. We optimize the communication data volume and hide much of the communication cost with local compute phases of the algorithms. Results show near-ideal scalability up to 1024 NVIDIA V100 GPUs on Summit, with performance exceeding 2.3 Tflop/s/GPU for the matrix-vector multiplication, and 670 Gflops/s/GPU for matrix compression, which involves batched QR and SVD operations. We illustrate the flexibility and efficiency of the library by solving a 2D variable diffusivity integral fractional diffusion problem with an algebraic multigrid-preconditioned Krylov solver and demonstrate scalability up to 16M degrees of freedom problems on 64 GPUs.

cs.DC↗

arXiv2020

High Performance Multivariate Geospatial Statistics on Manycore Systems

Modeling and inferring spatial relationships and predicting missing values of environmental data are some of the main tasks of geospatial statisticians. These routine tasks are accomplished using multivariate geospatial models and the cokriging technique. The latter requires the evaluation of the expensive Gaussian log-likelihood function, which has impeded the adoption of multivariate geospatial models for large multivariate spatial datasets. However, this large-scale cokriging challenge provides a fertile ground for supercomputing implementations for the geospatial statistics community as it is paramount to scale computational capability to match the growth in environmental data coming from the widespread use of different data collection technologies. In this paper, we develop and deploy large-scale multivariate spatial modeling and inference on parallel hardware architectures. To tackle the increasing complexity in matrix operations and the massive concurrency in parallel systems, we leverage low-rank matrix approximation techniques with task-based programming models and schedule the asynchronous computational tasks using a dynamic runtime system. The proposed framework provides both the dense and the approximated computations of the Gaussian log-likelihood function. It demonstrates accuracy robustness and performance scalability on a variety of computer systems. Using both synthetic and real datasets, the low-rank matrix approximation shows better performance compared to exact computation, while preserving the application requirements in both parameter estimation and prediction accuracy. We also propose a novel algorithm to assess the prediction accuracy after the online parameter estimation. The algorithm quantifies prediction performance and provides a benchmark for measuring the efficiency and accuracy of several approximation techniques in multivariate spatial modeling.

cs.DC↗

arXiv2020

Exploiting Low Rank Covariance Structures for Computing High-Dimensional Normal and Student-$t$ Probabilities

We present a preconditioned Monte Carlo method for computing high-dimensional multivariate normal and Student-$t$ probabilities arising in spatial statistics. The approach combines a tile-low-rank representation of covariance matrices with a block-reordering scheme for efficient Quasi-Monte Carlo simulation. The tile-low-rank representation decomposes the high-dimensional problem into many diagonal-block-size problems and low-rank connections. The block-reordering scheme reorders between and within the diagonal blocks to reduce the impact of integration variables from right to left, thus improving the Monte Carlo convergence rate. Simulations up to dimension $65{,}536$ suggest that the new method can improve the run time by an order of magnitude compared with the non-reordered tile-low-rank Quasi-Monte Carlo method and two orders of magnitude compared with the dense Quasi-Monte Carlo method. Our method also forms a strong substitute for the approximate conditioning methods as a more robust estimation with error guarantees. An application study to wind stochastic generators is provided to illustrate that the new computational method makes the maximum likelihood estimation feasible for high-dimensional skew-normal random fields.

stat.CO↗